Skip to main content

Full text of "Journal of the Society of Motion Picture Engineers"

See other formats


From the collection of the 



> 

Prelinger 
v o c jjit)rary 

v/ 

t P 



San Francisco, California 
2007 



JOURNAL OF THE SOCIETY OF 
MOTION PICTURE ENGINEERS 

VOLUME XXXVIII JANUARY, 1942 

CONTENTS 

The IR System: An Optical Method for Increasing 
Depth of Field ALFRED N. GOLDSMITH 3 

A New Dichroic Reflector and Its Application to Photo- 
cell Monitoring Systems G. L. DIMMICK 36 

Production and Release Applications of Fine-Grain 
Films for Variable-Density Sound-Recording 

C. R. DAILY AND I. M. CHAMBERS 45 

Laboratory Modification and Procedure in Connection 
with Fine-Grain Release Printing 

J. R. WILKINSON AND F. L. EICH 56 

A Note on the Processing of Eastman 1302 Fine Grain 
Release Positive in Hollywood V. C. SHANER 66 

Report of the Theater Engineering Committee 74 

Report of the Standards Committee 87 

Some Equipment Problems of the Direct 16-Mm Pro- 
ducer L. THOMPSON 89 

Current Literature 103 

Society Announcements 104 



(The Society is not responsible for statements oj authors.) 



JOURNAL OF THE SOCIETY OF 
MOTION PICTURE ENGINEERS 

SYLVAN HARRIS, EDITOR 

Board of Editors 

ARTHUR C. DOWNES, Chairman 

JOHN I. CRABTREE ALFRED N. GOLDSMITH EDWARD W. KELLOGG 

CLYDE R. KEITH ALAN M. GUNDELFINGER CARLETON R. SAWYER 

ARTHUR C. HARDY 

Officers of the Society 

* President: EMERY HUSE, 6706 Santa Monica Blvd., Hollywood, Calif. 
* 'Past-President: E. ALLAN WILLIFORD, 30 E. 42nd St., New York, 
N. Y. 

* Executive Vice-President: HERBERT GRIFFIN, 90 Gold St., New York, 

N. Y. 
^Engineering Vice-President: DONALD E. HYNDMAN, 

350 Madison Ave., New York. N. Y. 
^Editorial Vice-President: ARTHUR C. DOWNES, 

Box 6087, Cleveland, Ohio. 

* Financial Vice-President: ARTHURS. DICKINSON, 28 W. 44th St., New 

York, N. Y. 

* Convention Vice-President: WILLIAM C. KUNZMANN, 

Box 6087, Cleveland, Ohio 

*Secrelary: PAUL J. LARSEN, 44 Beverly Rd., Summit, N. J. 
^Treasurer: GEORGE FRIEDL, JR., 90 Gold St., New York, N. Y. 

Governors 

*MAX C. BATSEL, 501 N. LaSalle St., Indianapolis, Ind. 
**FRANK E. CARLSON, Nela Park, Cleveland, Ohio. 

*JOHN G. FRAYNE, 6601 Romaine St., Hollywood, Calif. 

*ALFRED N. GOLDSMITH, 580 Fifth Ave., New York, N. Y. 
**EDWARD M. HONAN, 6601 Romaine St., Hollywood, Calif. 
**JOHN A. MAURER, 117 E. 24th St., New York, N. Y. 

*LORIN L. RIDER, 5451 Marathon St., Hollywood, Calif. 



* Term expires December 31, 1942. 
** Term expires December 31, 1943. 



Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, included 
in their annual membership dues; single copies, $1.00. A discount on subscription or single 
copies of 15 per cent is allowed to accredited agencies. Order from the Society of Motion 
Picture Engineers, Inc., 20th and Northampton Sts., Easton, Pa., or Hotel Pennsylvania, New 
York, N. Y. 

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 

General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 

Entered as second-class matter January 15, 1930, at the Post Office at Easton, 

Pa., under the Act of March 3, 1879. Copyrighted, 1942, by the Society of Motion 

Picture Engineers, Inc. 



THE IR SYSTEM: AN OPTICAL METHOD FOR INCREASING 
DEPTH OF FIELD* 



ALFRED N. GOLDSMITH** 



Summary. The depth of field of a corrected lens system is determined by its 
focal length, its relative aperture, and the permissible diameter of the "in-focus" 
image of a point source. 

The limited depth of field in motion picture photography restricts freedom of action 
in large parts of the set, dictates a stylized, protracted, and costly studio procedure, 
and affects the dramatic value and audience appeal of monochrome or color pictures. 

Previous attempts to increase lens depth of field have been scientifically unsound 
and unsuccessful in practice. The new IR System, described in this paper, is based 
on a division of the set into optically appropriate regions, each region having identifi- 
able illumination, with the identification and differential focusing at the camera of all 
regional images within a single exposure. Thus greatly increased depth of field in 
straight and process shots becomes available. 

By a further modification in regional lighting, a number of such, increased-depth 
takes can be simultaneously made from different distances or at various angles, each 
such take having its own different and appropriate lighting. 

1. INTRODUCTION 

The following paper is submitted as a report of progress made by 
the staff of IR System, f In it there will be described the solution of 
a long-standing problem in the field of optics, namely: the attain- 
ment of greater depth of field than is obtainable by any previous 
method of utilizing a lens system for image formation. The solution 
is particularly applicable in the fields of photography and television 
under conditions of -controllable lighting of the external objects to be 
depicted. In this paper there will also be included methods for dem- 
onstrating the correctness and effectiveness in practice of the in- 
creased-range system which, as stated, has been invented to meet the 
need for increased depth of field, as well as indications of certain of 

* Presented at the 1941 Fall Meeting at New York, N. Y.; received Septem- 
ber 18, 1941. 

** Consulting Engineer, New York, N. Y.; Past-President of the Society of 
Motion Picture Engineers. 

fAs described in part in United States patents numbers 2,244,687 and 
2,244,688. 



4 ALFRED N. GOLDSMITH [J. S M. P. E. 

the directions in which the practical evolution of the IR System may 
reasonably be expected to proceed under studio conditions. 



2. NEED FOR INCREASED DEPTH OF FIELD 

It is correctly axiomatic among optical experts that the depth of 
field of a lens is a function only of the focal length of the lens and its 
working or relative aperture (together with the permissible diameter 
of the "circle of confusion" forming the image of a point source). 
Thus, for a lens of any specific focal length and aperture (or "speed"), 
focused for a definite object distance, and for a given permissible size 
of the circle of confusion of the point image, the depth of field is 
readily calculable and is unchangeable by any known method. 
Thus every objective lens in effect divides the object space into two 
portions. One of these portions is a relatively thin slab, based on the 
focused plane, and extending optic-axially forward and backward to 
an extent determined by the depth of field. Within this slab or re- 
gion, all objects are said to be depicted "in focus" on the conjugate 
plane of the photographic film or other photo-sensitive surface. The 
other portion comprises the entire remainder of the object space, 
and all objects within that second portion are depicted "out of focus." 
For object distances notably less than the hyperfocal distance of the 
lens, the second portion of space, wherein objects are incorrectly de- 
picted as out of focus, is vastly greater in extent than the first portion 
of space within which they are depicted in focus. This is a serious 
and inevitable limitation of all lens systems, and naturally has a pro- 
found influence on optical imagery in the motion picture field. Fur- 
ther, it influences and indeed controls the method of studio photog- 
raphy, script writing, separation of and relationship between various 
takes, editing, cutting, and the final effect of the produced picture. 
It has required the cameraman as well as the writer, director, actor, 
and editor to accommodate themselves to the rigid framework of a 
cramping optical law. Sincere tribute must be paid to the effort and 
ingenuity of these workers in their attempt to circumvent or at least 
to minimize the difficulties of photography within these restrictions. 
But, at best, an artificially limited presentation has resulted, includ- 
ing a number of conventions which have been accepted by the pro- 
ducers and the audience alike in the absence of anything better. 
Sometimes, indeed, cameramen have resorted more or less in des- 
peration to radical measures such as using extremely short-focus 



Jan 1942] THE IR SYSTEM 5 

lenses stopped far down in order to secure somewhat increased depth 
of field. In so doing, they have inevitably introduced the exagger- 
ated perspective and pictorial unnaturalness resulting from the use 
of such lenses, and have either over-illuminated the stage with re- 
sulting eyestrain and discomfort to the actors or have produced a con- 
trasty, underexposed, "chalk and coal" picture or both. 

On the other hand, the human eye has in effect a practically un- 
limited depth of field. The re-focusing of the eye for any desired ob- 
ject distance is so nearly instantaneous that the viewer is unaware 
that any part of the field of view is ever out of focus. Thus, in natural 
and real life, persons with normal sight are totally unaware of any 
limitation in the visual depth of field. Again, anyone who, in view- 
ing a legitimate stage performance, found that he could see the people 
in the foreground sharply defined but that the people in the back- 
ground were blurred or fuzzy, would at once and properly consult his 
oculist. 

The camera lens, by way of contrast, has a very limited depth of 
field. Considering the most generally used lens of 50-mm focal 
length operating at a speed of, say,//2.8, it is interesting to note that 
the back depth of field for an object distance of 5 feet is only 6 inches ; 
for an object distance of 8 feet is 1 foot 2 inches; and for an object 
distance of even 12 feet is only 3 feet 2 inches. It is within such cir- 
cumscribed regions that conventional photography must be carried 
out. The follow-focus adjustment is of course no solution of the 
problem since it applies only to one given object distance at any in- 
stant and does not in the least solve the major pictorial and dramatic 
problem of successfully depicting significant action occurring simul- 
taneously at widely different distances from the camera. 

Thus, in each take, the motion of the actor is definitely and neces- 
sarily limited by chalk lines on the floor. This regulates the nature 
of the action which can be included in the script; prevents the usual 
and natural action of various groups simultaneously at different dis- 
tances from the camera; forces the choppy method of frequent 
jumping back and forth between long shots, medium shots, and 
close-ups with all the resulting multiplicity of takes, and difficulties 
in maintaining mood and matched action (not to mention matched 
sound) ; and cannot but affect adversely the naturalness of acting for 
the motion picture as compared with that of the legitimate stage. 
It is appropriate briefly to quote certain major Hollywood figures in 
this regard : 



6 ALFRED N. GOLDSMITH [J. S. M. p. E. 

"Hal Mohr, when assigned to Green Pastures, decided the dramatic power of the 
photograpy would be considerably increased if, instead of breaking up the action 
between a figure in the foreground and another in the background into a series of 
individual close-ups, the action could be shot at the same time with one lens." 

This viewpoint favoring continuity of photography through the 
entire depth of the set is significant. The reaction of the actor to the 
restrictive effects of limited depth of field is indicated in the following 
quotation : 

"Hollywood, in the opinion of Eduardo Ciannelli, would be a paradise for the 
stage actor if it weren't for the chalk lines. . . . And he pointed to his feet, to the 
veritable chalk lines the screen player must toe. 'If it weren't for those,' Cian- 
nelli went on, 'which are there to keep you from walking out of the camera range 
or getting so close you're out of focus or so far away they can't see you, films would 
be for the stage player all he has ever dreamed of.' " 

The following comment by Allyn Joslyn expresses a similar thought 
in another form : 

". . . Finally it is my turn to get in front of the cameras. I can't turn my head 
too far to the left or I'll be out of line. I can't take a step too far to the right or 
I'll be out of focus. In other words, there I am, penned in a tiny space, realizing 
that unless I stay in it and give all I've got they will have to make a retake." 

The effect of this claustrophobic optical atmosphere on the actor is 
further graphically indicated by Paul Muni in the following : 

"To act in motion pictures is to act in a world in which mechanical problems 
beset the actor on all sides; his performance is governed by them, he can not 
escape them. From the time he appears on the set, his steps are caged by chalk 
marks, and focal distances . . . and his image can only be seen if he moves with 
care within the cage." 

Speaking of the difficulty experienced in economical set construction 
when panning shots are made under present conditions, Hans Dreier 
says: 

"A more involved example would be one in which the camera precedes an actor 
walking along, then swings away from him to point out a person in another part 
of the room. Such a set-up sometimes means the construction of large sets which 
are subsequently overlooked, because the camera is so close to the actor that 
everything else is out of focus." 

It is necessary to stress here the preceding points since many studio 
workers have become so accustomed to present-day practice that its 
limitations have come to be regarded as inevitable, natural, and in- 



Jan., 1942] THE IR SYSTEM 7 

herently to be accepted. Indeed, present motion picture technic is 
fundamentally based on these optical limitations of the objective 
lens. Studio practice simply does its best within the limits of present- 
day optics. Some cameramen and a part of the audiences have be- 
come so accustomed to the present restrictions that they hardly real- 
ize the wealth of improvement which would be available were these 
restrictions to be removed. It does not occur to most people that the 
present abrupt succession of long, medium, and close-up shots neces- 
sarily results in large measure from the limitations of the lens. Pres- 
ent practice requires continually shifting the point of view, weari- 
somely accumulating many takes of each scene or action, painstak- 
ingly matching up successive takes, "covering up" in cutting and 
editing, delay and increased cost in production, unnaturalness in the 
acting and in the corresponding effect on the audiences, and a less 
economic and dramatic set-up than would otherwise be attainable. 

The availability of increased depth of field would gradually bring 
about a marked revolutionary change in methods of production, 
greater flexibility, considerable economies, and simplifications in 
camera technic. To the audience, the pictures would more closely 
resemble legitimate-stage performances or even real life. It may be 
added that conventional pictures of today, with their limited depth 
of field, cause an unconscious irritation or strain to the audience. 
While the foreground may be in focus, the background is usually 
blurred. In accordance with long custom and experience, the eye of 
the viewer attempts to bring the background into sharp focus and of 
course fails. This continual attempt and failure is a physiological 
strain and a psychological disappointment. 

In the case of color motion pictures, increased depth of field is, if 
possible, even more essential. Color pictures, when out of focus, are 
not merely blurred but also "out of color." Thus, a red-and-white 
checked dress when out of focus is merely pink which is something 
quite different from the original. An out-of -focus background of 
colored objects usually shows curious bronzy greens, nondescript 
browns, and other unprepossessing and incomprehensible color effects. 

From the viewpoint of the film manufacturer, the limited depth of 
field now current somewhat nullifies the advantages sought in the 
painstaking production of fine-grain film. The high possible "load 
factor" of the film is not realizable under present conditions. Sharp- 
ness of delineation is limited to a relatively small portion of the de- 
picted space, and thus most of the film might as well be extremely 



8 



ALFRED N. GOLDSMITH 



tf. S. M. P. E. 



coarse-grained so far as any picture-carrying capabilities are con- 
cerned. 

In Fig. 1 is shown schematically the diameter of the circle of con- 
fusion of the image of a point source at various distances from the 
lens (dotted curve), together with the equivalent resolution, expressed 
in terms of "lines" comprised in the height of the picture (these being 
adjacent lines like those used for scanning in television practice) 
(solid curve). There is also shown the "useful" definition realized in 
the picture, together with the "lost" definition. It is clear enough 




FIG. 1. Picture resolution versus distance, as controlled by depth of field. 
Focal length, 50-mm; aperture, //2. 3. 

that present practice constitutes an uneconomic combination of fine- 
grain film and coarse-grain images. 

It is assumed that a resolution of 70 lines and equally wide spaces 
between them per millimeter can be achieved in the focal plane, this 
corresponding to 2100 lines resolution according to the present tele- 
vision practice. It is not meant to imply that such resolutions are 
commercially attainable at all times in the negative nor that they can 
be maintained in projection on the theater screen. However, they 
represent a possible delineatory capability of the film. The size of the 
image of a point source on the film is indicated for various distances 



Jan., 1942] THE IR SYSTEM 9 

from the lens on the vertical lines marking those distances. The 
corresponding approximate resolution is indicated on the curve. It 
is obvious that, except for one sharp maximum, the resolution of the 
picture soon falls to inacceptable values and indeed to values re- 
jected in the past as inadequate for television service in the home. 
Outside of the region embraced by the narrow maximum of resolu- 
tion in question, the capabilities of the film, the script, and the actors 
are unrealizable on the theater screen. 

Another fundamental need for increased depth of field will be in the 
domain of three-dimensional pictures. No matter what type of 
stereoscopy is used in the future for that purpose, it is obvious that 
for naturalness of the three-dimensional pictures the depth of field 
must appreciably cover the depth of the object space to be depicted. 
There is no naturalness to "three-dimensional fog banks," so to speak 
and that is all that can be produced in the out-of -focus background 
of a stereoscopic picture. 

Once increased depth of field becomes available, there will natu- 
rally be a demand for a method of making simultaneous multiple angle 
shots. That is, it will be preferred to shoot the same action simul- 
taneously from various viewpoints and distances, with all parts of the 
picture in focus, and with each camera viewing the action illuminated 
by a particular type 'of lighting appropriate to the effect which is de- 
sired. Further operating economy and naturalness in the acting and 
resulting picture would thus be attained. The editor and cutter are 
given a far wider opportunity to do a masterly job with the avail- 
ability of multiple-angle-shot negatives of the same action, taken 
simultaneously, and in focus throughout. 

It was with such major considerations in mind, that the attempt 
was made by the staff of IR System to solve the corresponding prob- 
lems. At first sight, the task seemed sufficiently discouraging. 

3. HISTORY OF PREVIOUS ATTEMPTS TO INCREASE DEPTH OF FIELD 

Numerous unsuccessful attempts in the past have sufficiently dem- 
onstrated the difficulty of the problem, as well as its importance. 
Previously suggested methods have all contained more or less subtle 
and concealed fallacies. As a natural consequence, attempts to solve 
the problem of depth of field had acquired something of the low re- 
pute of "perpetual motion" inventions in the minds of most reputable 
scientists. The point had been reached where anyone proposing an 
increased depth-of-field method had been more or less automatically 



10 ALFRED N. GOLDSMITH [J. S. M. p. E 

classified as flying in the face of scientific facts, laws, and limitations. 
And it must be admitted that the previous attempts in this direction 
had largely justified this classification. 

Broadly, all previous methods were based on a degradation of 
image quality. That is, if the permissible diameter of the (circle of 
confusion of the) image of a point source in its conjugate focal plane 
is increased, depth of field is also increased. But the problem has 
not in the least been solved. It has not even been evaded. Poor 
photographic quality and unsharp images are obviously too high a 
price to pay for depth of field. The problem instead is just this: 
to increase depth of field while maintaining image sharpness and 
photographic quality. Contrariwise, the previous attempts can be 
readily classified on the basis of the method selected for degrading the 
image quality. None of them solved the depth-of -field problem with 
maintained pictorial quality. 

The two chief sub-groups of these image-quality-degradation 
methods are the following. Both are theoretically incorrect and 
physically unworkable and inoperative. They are here mentioned 
for the sake of completeness. 

Group 1. Aberrational Pictures. In this method, optical aberra- 
tions of some sort are deliberately or unconsciously introduced into 
the objective lens system. Usually spherical aberration or zonal 
aberration or both have been introduced, although comatic (and 
even chromatic) aberrations have also been included. Sometimes 
the aberrations have been introduced in the design or construction 
of the lens, and sometimes they have been added by placing an addi- 
tional aberration-producing component lens in front of the main 
objective. 

The resulting images were not sharp or clear anywhere. They 
were no solution of the depth-of -field problem. Increased depth of 
field in the true sense does not result from looking at pictures through 
a fog, a haze, or a diffusing gauze. 

Group 2. Average-Focus Pictures. In this method, the entire set 
from foreground to background is illuminated steadily. The lens is 
moved axially (actually or in effect) during the exposure of each 
frame so that it is focused on the foreground during part of the expo- 
sure and on the background during another part of the exposure. 
The fallacious and unplausible claim is made that, during the portion 
of the exposure that the foreground is in focus, it is sharply photo- 
graphed ; and similarly during the portion of the exposure when the 



Jan., 1942] THE IR SYSTEM 11 

background is in focus, it is photographed in focus; and, therefore, 
add the advocates of this method, the resulting picture is completely 
in focus! 

The facts are quite otherwise. While the lens is focused on the 
foreground, the out-of-focus background is also photographed since 
there is nothing discriminatory about the lens or the film. Likewise, 
when the lens is focused on the background, the out-of-focus fore- 
ground is photographed. As a result, the images of objects at any 
and every distance from the lens consist of an in-focus component 
overlaid and badly blurred by an infinity of out-of-focus pictures of 
the objects at that distance from the lens. Obviously, the combina- 
tion of an in-focus picture with an infinity of out-of-focus pictures is 
a blurred picture in every plane, and not a sharp picture as claimed. 

The fallacy in this method is as evident as the following. A tea- 
spoonful of pure water is mixed with many teaspoonfuls of water 
each containing increased proportions of a deadly poison. Will the 
mixture be pure water, or will it be poisonous? Substitute for "pure 
water" a sharply focused component picture, and substitute for 
"poison" an out-of-focus component. It then becomes clear enough 
that the average-focus method is merely another way of trading image 
quality for depth of. field, and is therefore totally inaccep table. 

As indicated above, attempts have also been made to increase 
depth of field by using very short-focus lenses greatly stopped down. 
This method is theoretically correct within its limits, but it is useless 
for acceptable photography. The requirements for correct perspec- 
tive in the projected picture are well known and not to be slighted. 
The short-focus lens in the camera gives pictures which, when pro- 
jected in the theater and observed at normal viewing distances, are 
badly distorted and show exaggerated perspective. Thus the re- 
quirements for correct perspective are shockingly violated in this type 
of picture. Further, motion of the actors is intolerably peculiar with 
strange variations in velocity which are utterly unlike those in real 
life. Then, too, the stopped-down lens leads to vastly increased and 
obviously inacceptable studio lighting for a nearly correctly exposed 
picture or, alternatively, to considerably increased and just tolerable 
studio lighting with an underexposed and contrasty negative. 

In the problem of depth of field, as in some other human problems, 
"we thus learn from history that we learn nothing from history" 
except perhaps what to avoid. 



12 ALFRED N. GOLDSMITH [J. S. M. p. E. 

4. SPECIFICATIONS FOR AN INCREASED-RANGE SYSTEM 

In view of the confusion which has existed in some quarters as to 
what constitutes a system for true increase in depth of field, it is 
deemed appropriate to list below the specifications and general char- 
acteristics of such a system. The following description then covers 
an acceptable method for adding to depth of field without optical or 
esthetic deterioration of the picture rather than a system which 
merely sacrifices important optical or pictorial qualities to secure an 
apparent increase in depth of field. The following are, however, ad- 
mittedly strict specifications. 

(a) The system shall be usable for either black-and-white or color pictures. 
(6) It shall be usable for still-picture or motion-picture photography. 

(c) It shall be usable for photography or for television and the like. 

(d) The system shall enable using standard lenses of the highest degree of cor- 
rection of optical aberrations then current in the art. 

(e) It shall enable using lenses of any normal and usual focal lengths. 
(/) It shall enable the use of normal lenses at the usual large apertures. 

(g) And above all, there shall be no deterioration of picture quality or sharpness 
as the result of the use of the system. 

(h) The sharpness of focus of objects in any portion of the object space shall be 
reasonably controllable. This condition, which is preferred but not mandatory, 
is met by the IR System. 

(i) No extensive or unduly cumbersome changes in camera construction shall 
be necessary. 

(j) The handling of the camera and its finder shall be essentially conventional; 
and increased-range pictures shall be available in the finder to guide the camera- 
man as well. 

(k) It shall require no more total light on the actors than for ordinary photog- 
raphy. 

(/) It shall enable the use of lighting to obtain any of the effects obtained by 
conventional lighting and photography. 

(m) Setting up the lighting of the set shall require little specialized optical 
knowledge, and shall be conveniently possible by the usual experienced electricians. 

() It shall be flexibly adaptable to composite or process shots as well as to 
miniature work. 

(0) It shall preferably enable making a number of simultaneous angle shots from 
different viewpoints, with different types of lighting; and each of these shall have 
increased depth of field as required. 

While the above are extremely difficult specifications, they have 
been met by the newly developed IR System. So far as is known, 
they have never previously been met; and there is reason to believe 
that no other method of meeting these specifications with conven- 
tional optical elements alone is possible. 



Jan., 1942] THE IR SYSTEM 13 

5. GENERAL PRINCIPLES OF THE IR SYSTEM 

In ordinary photography, each frame or picture represents a view 
from a uniquely focused lens of an entire and overall-illuminated ob- 
ject space. In the IR System, each frame or picture represents the 
composite of a number of views made by a multiply focused lens, 
each of which views covers only a region or division of the object 
space so selected that the picture of such region is altogether in focus. 
Thus the IR System includes the photography within a single expo- 
sure period (single frame) of the various adjoining regions of the 
object-space in such fashion that 

(a) each region has an identifiable illumination, with minimized 
spillover of illumination of other types, and 

(b) the identifiable illumination from each region is first identified 
and in effect segregated, and; second, brought to the same focal plane 
as that from all other regions by means of a differential-focusing de- 
vice. 

The identifiable regional illumination may be of the following types : 

(a) It may be suitably polarized light. This method is not recom- 
mended in general because of the non-retention of the plane of polari- 
zation of light on reflection from most surfaces. 

(b) It may be appropriately colored light. In this case, each re- 
gion in the object space is illuminated by light occupying a different 
spectral range, thus enabling its later identification either by color 
filters or by the differential refraction (dispersion) of some refracting 
system. The method amounts to establishing a selected color con- 
figuration (e. g.y a spectrum, broadly speaking) axially through the 
object space, and then introducing a compensating amount of longi- 
tudinal chromatic aberration into the objective lens system so that 
the colored light from each region is brought to the same focal plane 
and there correctly imaged. 

(c) It may be correctly timed illumination. In this instance, the 
total time for each frame exposure is divided into a number of por- 
tions. Each region in the object space, as defined below, is illumi- 
nated during a selected portion or portions of the total available expo- 
sure time and is not illuminated at any other times during the exposure 
period. This last proviso is fundamentally novel, and is important in 
the realization of the method. At the camera lens, a differential- 
focusing device is synchronized with the light coming from each 
corresponding region in such fashion that the light from each region is 
brought to the same focal plane. The resulting picture therefore has 



14 ALFRED N. GOLDSMITH [J. S. M. p. E. 

increased depth of field since it consists in effect of the composite of a 
number of pictures each of which is inherently in focus, which pic- 
tures in their totality form a complete picture of the object space. 

This method amounts, in its simplest form, to sending a wave of 
illumination (in the form of an axially extensive and steadily deepen- 
ing slab of illumination) in some systematic fashion through the ob- 
ject space, and pacing it by a synchronized wave, so to speak, of 
correlated focusing of the objective lens such that the light from all 
objects on the set is brought to the same focal plane. It also amounts 
to a sort of third-dimensional luminous scanning of the set followed 
synchronously by the appropriately coordinated focusing of the lens 
so that the picture of the entire set is in focus on the film. 

Methods a, b, and c above can be combined in various ways. It 
is not necessary here to go into details in that regard. 

Method b as described above can be used only for black-and-white 
or other monochrome photography. It is interesting in that there 
are no moving parts involved. Method c will generally involve at 
least some moving parts. It can be used for either black-and-white 
or color photography. 

Method c, as suggested above, is complementary or supplementary 
to the usual two-dimensional television scanning. It is in fact a form 
of planar scanning in the third dimension (that is, parallel to the 
optical axis of the objective lens) . 

Inasmuch as the focusing of the images of objects lying within any 
region in the set is dependent on the identifiable illumination of that 
region, it is clear that the image of each region can be controllably fo- 
cused or softened to any desired extent regardless of the sharpness of 
focus of other regions. Thus individual objects can be changed in 
sharpness by altering or destroying the identifiable characteristic of 
the light falling on such objects. Accordingly, it is possible to pro- 
duce what would otherwise seem to be an optical anomaly, namely, 
and as an example, pictures in which the foreground and background 
are sharply focused but in which the middleground is softened to any 
desired extent. The imaginative director and cameraman will read- 
ily see numerous dramatic and comic possibilities in so versatile a 
type of photography. 

6. DEFINITIONS OF IR SYSTEM TERMS 

Multi-regional photography by the IR System is so novel a method 
of optical imagery that it naturally leads to some new optical con- 



Jan., 1942] THE IR SYSTEM 15 

cepts which, for convenience, should be given appropriate names and 
denned. This has been attempted in the following set of definitions 
of terms which have been found conveniently usable by the workers 
in that field. The list is incomplete since a considerable number of 
the concepts and methods involved in the IR System are beyond the 
scope of this paper, and accordingly the terms involved in those rela- 
tions are not here included. 

Region. A portion of space symmetrical about the optical axis of 
an objective lens, based upon a central plane of sharp focus, and 
bounded externally front and rear by surfaces corresponding to a pre- 
determined and maximum tolerable value of the diameter of the circle 
of confusion of the image of a point source placed upon such surfaces. 

Division. A portion of space symmetrical about the optical axis 
of an objective lens, based upon a central plane of sharp focus, and 
bounded externally front and rear by surfaces corresponding to a pre- 
determined value of the diameter of the circle of confusion of the 
image of a point source on such surfaces, such value being less than 
the tolerable maximum. A division thus resembles a region but cor- 
responds to less than the full depth of focus of the lens. Thus, there 
will be more divisions in a given object space than there are regions. 

Differential Focuser (briefly, Diffo). An optical element placed 
suitably in relation to an objective lens and capable of rapidly shifting 
the back focal plane thereof according to a predetermined time sched- 
ule during a picture-making period, but without the introduction of 
perceptible aberrational errors in the conjoint optical system. 

Space Spillover. The amount of light intended for illumination of 
one region (see above) which actually reaches another region (ex- 
pressed in percentage of the light normally sent into the second re- 
gion). The definition can also be applied to divisional space spill- 
over. 

Time Spillover. The amount of light intended for illumination of 
a given region within a given period which begins before or lasts be- 
yond such period of illumination (expressed in percentage of the nor- 
mal illumination of that region during that period). This definition 
may also be applied to divisional time spillover. 

Space Registration. The deviation from exact registration during a 
complete exposure between the outer edges of opaque objects in a 
frontal region and the inner edges of the thereby-occulted or masked 
outlines in a region further removed from the lens. 

Motional Registration. The deviation from exact registration dur- 



16 



ALFRED N. GOLDSMITH 



[J. S. M. p. E 



ing a complete exposure between the positions or outer edges of mov- 
ing objects in a frontal region and the inner edges of the thereby- 
occulted outlines in a region further removed from the lens. 

For convenience of reference, it may be here added that space reg- 
istration effects result from the relatively unfocused edges of frontal- 
region objects, acting as masks during the photography of further re- 
moved regions. Motional registration effects, on the other hand, re- 
sult from the shifted edges of frontal-region objects in motion at the 
time of photographing further removed regions. 

7. OPERATION OB IR SYSTEM IN STUDIO 

In order to make the method of operation of the timed-illumination 
form of the IR System clear, a schematic arrangement is shown in 




FIG. 2. 



Regional illumination and camera arrangements of a set according 
to the IR System. 



Fig. 2 illustrating the illumination of the set. The central planes of 
the regions are indicated, but it should be remembered that these 
planes are somewhat forward of the geometrical center of the region 
since the back depth of field always exceeds the front depth of field. 
The front and back boundaries of the regions are also shown. The 
regional illumination in each case is seen to be directed into that re- 
gion only, and with minimum space and time spillover. The cam- 
era shutter is replaced by a differential focuser or diffo, which oper- 
ates synchronously with the flashing of the lights in each region. 
This changes the position of the focal plane systematically so that 
each region is imaged in the same plane, namely, that of the film. 



Jan., 1942] 



THE IR SYSTEM 



17 



Considering Region 1, F.P. 1 is the central or focal plane. Top 
lighting into the region is indicated by TL 1; side lighting by the 
oblique lamp SL 1; bottom lighting is indicated by the lamp BL 1; 
and back lighting for Region 1 by the lamp BAL 1. Similarly top 
lighting for Region 2 is indicated by TL 2, and for Region 3, by TL 3. 
The coordination or synchronizing of the illumination from these 
various lamps with the angular position of the diffo in the camera is 
schematically indicated by the dashed lines in Fig. 2. 

It is obvious that there are many ways of focusing differentially 
and of timing the regional illuminations in synchronism with the phase 
or position of the diffo. Typical examples of these will be described 
further below. 



O.0020 



O.0076 




O.0012 



0.0003 



O.OO04 



\$4 



FIG. 3. Diameter of point-source image versus distance for IR System opera- 
tion. For lens of focal length of 50-mm and stop//2.3. 

The system shown in Fig. 2 leads to an interesting form of the 
curve connecting distance from the lens and diameter of the circle of 
confusion in the image of a point source at the corresponding distance. 
This multi-cusped curve is shown in Fig. 3. The positions of the 
central planes of the regions, the regional boundaries, and the maxi- 
mum permissible circle-of-confusion diameter (taken as 0.002 inch for 
35-mm film) are all indicated on the drawing. The left-hand loop 
corresponds to ordinary photography, while the complete curve corre- 
sponds to the increased depth of IR System photography. 

By way of comparison, there is also shown on Fig. 3, as a dot- 
dashed curve, the depth of field of a lens of the same focal length 



18 ALFRED N. GOLDSMITH [j. s. M. p. E. 

stopped down to the point where it covers the same total or overall 
depth of field as three-regional IR System photography. It will be 
seen that the results of the two methods are even then by no means 
identical, or indeed equivalent in photographic effect, that is, even 
when the stopped-down lens is used to obtain equal overall depth. 
The IR System curve shows clearly three minima, corresponding to 
planes of sharpest or "absolute" focus. Thus, it is quite possible to 
place the central or most sharply focused plane of the rearmost Re- 
gion 3 in the IR System at the back wall of the set or, alternatively, on 
the screen forming the back drop for rear projection composite shots 
(as described more fully below). In this way, the important back- 
ground is shown in completely sharp focus, as are also objects at sev- 
eral other distances from the lens and, of course, all other objects in 
the field are depicted "in focus" that is, within the acceptable limits 
of sharpness. In the case of the stopped-down lens, on the other 
hand, only objects at a single distance from the lens can be in absolute 
focus; and the important background, for example, is less sharp 
than this, although perhaps acceptable. 

Using IR System photography, it is also feasible to divide the ob- 
ject space into more divisions than the number of optical regions it 
comprises. In that case, the curve corresponding to Fig. 3 will have 
as many minima corresponding to absolute focus as there are divi- 
sions, and thus there will be more minima than there are regions. 
Furthermore, in this arrangement, the maximum value of the diame- 
ter of the point-source image will always be less than the permissible 
maximum of 0.002 inch for 35-mm film, and this diameter and the 
corresponding sharpness are in fact under control with the arrange- 
ment in question. 

Up to this point it has been tacitly inferred that the boundary sur- 
faces of the various regions in IR System photography are plane sur- 
faces perpendicular to the optical axis of the lens, as schematically 
indicated in Fig. 2. Such, however, is not the case. It is well known 
that the speed of a lens is always less for the edge of the field than for 
the center because, in the main, of the occulting effect of the lens 
tube and the interior stops for oblique passage of image-forming rays 
through the system. Thus the depth of field at the edges is greater 
than at the center. In consequence, the boundary surfaces of the 
regions are as indicated in Fig. 4. The central or focal plane of Re- 
gion 1 is indicated by F.P. 1, and the dotted lines Fl and Bl are sec- 
tions of the surfaces of revolution comprising, respectively, the front 



Jan., 1942] 



THE IR SYSTEM 



19 



and rear boundaries of Region 1. Similarly F.P. 2 is the central 
plane or Region 2, and the dot-dashed lines F2 and B2 represent the 
front and rear boundaries of Region 2 ; while F3 is the front boundary 
of Region 3, of which F.P. 3 is the central or focal plane. 

It is characteristic of the IR System that the illumination of any 
one region shall be restricted with minimum spillover to that region. 
From Fig. 4, it is obvious that oblique regional lighting in Region 1 
can enter the region at considerable angles with the optic axis of the 
camera lens without violating the condition in question. Increasing 
obliquities are available for the illumination in the successively fur- 
ther removed regions, as indicated by O.R.L. 2 and O.R.L. 3, respec- 
tively. A similar comment as to permissible obliquity of back light- 
ing applies, as indicated by the back oblique regional light B.O.R.L. 1 




FIG. 4. Boundaries and overlap of regions according to IR System, consider- 
ing center and edge speeds of lens. 

for Region 1 and by B.O.R.L. 2 for Region 2, respectively. As a re- 
sult of these flexible conditions, it proves to be conveniently possible 
to duplicate effectively any desirable light vector (that is, magnitude 
and direction of luminous flux) in each region. It is also interesting 
to note that the cross-hatched volumes of which the cross section are 
indicated as 1,2 and 2,3, respectively, in Fig. 4 can be appropriately 
illuminated according to the IR System requirements by light from 
either Region 1 or Region 2 for the volume 1,2 and by light from 
either Region 2 or Region 3 for the volume 2,3. This adds to the con- 
venience of regional lighting. 

During the development of the IR System, comprehensive tables 
were prepared giving the regional central distances and forward and 



20 



ALFRED N. GOLDSMITH 



[J. S. M. p. E. 



back limiting distances for an adequate variety of focal lengths of 
lenses used in motion picture work and also in television, for a con- 
siderable number of lens speeds, and for numerous working conditions 
covering an extremely wide variety of requisite depths of field. In 
their totality, these tables substantially eliminate any need for com- 
putation, reference to other depth-of-field tables, or delay in the ap- 
plication of the IR System to any desired circumstances and needs. 
Using these tables and certain supplementary data, it has been re- 
garded as interesting to determine the available "universal-focus" 











































. 






V 


F SC 






\ 










\ 
















X 






\ 


; 






\ 


^r 


SOtn 


77. 






\1 A7 ' 














\ 






\ 








\ 












\ 














\f 


40m 

1 


m.; 


v 


\ 






\ 


\ 












\ 














Y, 






\ 








\ 














X 












A 








\ 








\ 








/ 






^^>, 


^ 




3 


/ 

rfo/t/ 
f-<//?t 
BCK/ 


it 

Yf/>0 

y w/ij 


\ 






s 


s 






\ 


\ 




.*' 














sr\ 


\ 






\ 








\ 

s' 


>s 
























\ 








\ 




y' 




X 


X 
























\ 






f , 


X. 










^*^ 


^>^ 


^, 




















\ 


.--' 






^^"^* 


^ 


^ 










""*** 












A 
#f/U 


feoiit 

170* 


M, '^ 
MffS 


2 


\ 












^ 


-^ 


^ 


t 


MFD 

Vii / 


MM, 

CHGT 


v 






















I 


r 








s 


3 






y 



0/STA#C Of FOCi/SSf0 FOjeOfCH/#O 



FIG. 5. ,Universal-focus distances versus apertures for current IR System 

practice. 

capabilities of the IR System as applied in one very simple embodi- 
ment presently available. These data are shown in Fig. 5, but it 
should be kept in mind that by applying certain refinements and 
elaborations of the IR System, the performance shown in Fig. 5 can 
be considerably exceeded. The full-line curves in Fig. 5 show the 
foreground distance for each focal length and indicated stop for which 
the IR System will give back depth to infinity that is, universal 
focus. A comparison of these curves with the hyperfocal distances 
of the same lens at the same stop will show the magnitude of the in- 
crease in depth of field. Thus, for a 50-mm lens at //2.7, the avail- 
able universal depth is from a little over 10 feet to infinity for the IR 



Jan., 1942] 



THE IR SYSTEM 



21 



System, whereas the hyperfocal distance of the same lens at this stop 
is over 60 feet! 

For convenience, there have also been shown on Fig. 5 the dotted 
curves which intersect the full-lined curves at points corresponding 
approximately to photography of a "medium close-up, head nearly to 
waist" and "medium, head to knees," respectively. This is highly 
instructive to the practical director of photography and cameraman 
in connection with the dramatic capabilities of a photographic sys- 
tem which gives universal focus with reasonable lens stops from me- 
dium close-ups to medium shots, as the case may be. It may be 




FIG. 6. 



Optical basis of one form of diffo used in IR 
System. 



added that under conditions where a one-shot in a close-up can be 
made only with some difficulty by ordinary photography, the IR 
System will permit making two-shots or even three-shots with the 
several actors back of each other at acceptable and convenient separa- 
tions. 

As was mentioned above, there are a number of available methods 
for the differential focusing of the lens in synchronism with the 
sequential lighting of the corresponding regions in the object space. 
The basis of one form of differential focuser (diffo) is shown schemati- 
cally in Fig. 6. The lens is indicated in each case by the vertical line 
through its center. Assume that 0\ is an object in Region 1 and that 



22 ALFRED N. GOLDSMITH tf. S. M. P. E. 

the corresponding image is located at /i. An object 2 in Region 2 
produces an image at 7 2 closer to the lens. A diffo plate may be inter- 
posed in the path of the image-forming rays on their way to 7 2 . It is 
well known that the emergent rays in this case will be parallel to the 
incident rays but that the new image location /i will be displaced 
away from the lens by an amount depending upon the thickness of 
the plane-parallel diffo plate and its refractive index. By an ap- 
propriate selection of these constants, the image of 2 can be brought 
to the same location /i as was the case for the object 0\. If the thick- 
ness of the diffo plate is less than a certain fraction of the focal length 
of the lens, no discernible optical errors will be introduced into the 
thereby displaced image as a result of the interposition of the diffo 
plate in the path of the image-forming rays. 

A diffo is shown in Fig. 7 together with a corresponding aperture 
plate, both intended to replace the shutter in a certain standard form 
of studio camera, and without any modification of the camera mecha- 
nism or construction. The five diffo plates are visible in the right-hand 
portion of the figure, the thickest of these corresponding to the 
farthest region or background. 

The procedure in using the IR System under studio conditions may 
be briefly summarized in the following. This procedure markedly re- 
sembles that for ordinary photography. It is first determined what 
depth of field is desired, and also the distance of the foreground plane 
(that is, the nearest distance to the camera at which a person in 
absolute focus is to be located). The pre-computed tables of the IR 
System will then indicate for each focal length of lens and for each 
desired stop, the number of regions that will be required to cover the 
desired depth of field. Further, these tables will give the axial distance 
from the lens of the front and rear boundaries of the regions in 
question. The edge boundaries can generally be approximately de- 
termined from these tables by a simple procedure which need not be 
here described. As an alternative method, of a more conservative sort, 
it is merely necessary to determine the depth of field of a lens of the 
focal length which is desired when stopped down to an extent equal to 
the normal lens aperture multiplied by the number of regions. Thus, 
if a 50-mm lens operating at//2.8 is used for four-regional IR System 
photography, as a first and conservative approximation the working 
depth will correspond to that of a 50-mm lens operating at //1 1.2 
(and thus requiring sixteen times the previous amount of light on the 
the set!). 



Jan., 1942] 



THE IR SYSTEM 



23 




FIG. 7. Diffo and aperture plate for TR System camera operation 




FIG. 8. Flashing gas-lamp*illumination of miniature set according to IR 

System methods. 



24 ALFRED N. GOLDSMITH [J. S. M. p. E. 

Having thus determined the regional boundaries the lamps are set 
up to produce the desired key and modelling lighting for each region 
with minimum spillover into adjacent regions and reasonably good 
blending of lighting between regions. As a general rule this will 
merely require inserting the plug at the end of each lamp cable into a 
spider corresponding to the region which that lamp will illuminate. 
The power supply and adjunct wiring to the spider are of course 
properly associated with the power supply to the camera motor, 
both as to frequency and phase. 

The camera is set up at its appropriate location and checked, by 
means of the finder, for focus and depth of field. It may be added 
that the finder should preferably also be adapted to IR System 
viewing so that the cameraman has a correct idea of the depth of 
field in the corresponding picture. 

In the event that background projection is used, it becomes neces- 
sary to time the projection of the background so that it corresponds 
with the regional location of the projection screen. Otherwise stated 
the light-source in the projection lamp house has a phase and duration 
appropriate to the region in which the screen is located. Thus the 
background image is in complete focus thereby avoiding one of the 
most serious limitations in such composite pictures, namely, the con- 
flict in focusing, so to speak, between the actors in the foreground 
and the screen in the background. 

Experiments have been carried out to judge the amount of spillover 
light which is permissible. Experience has shown that the sharpness 
of photographic focus is not appreciably affected on important parts 
of each regional picture if the spillover does not exceed 3 per cent. On 
the most important planes it is usually preferred to keep the total 
spillover at 2 per cent or less. 

Extensive work with the IR System has made clear that the method 
is not a mathematical abstraction or a precision geometrical system 
under normal working conditions. It turns out to be a convenient 
method of operation with which liberties can be taken in practice 
and which can be used with the same latitude and discretion as any 
other photographic system. When first utilized, it is found strange to 
accept the IR System convention that light can not be poured 
through the set passing through one region after another but that 
the illumination must rather be handled sequentially and region- 
wise, so to speak. The duplication of the existing lighting effects by 
modifying lamp locations and directions seems at first unfamiliar. 



Jan., 1942] THE IR SYSTEM 25 

However, after a little experience with the system all strangeness dis- 
appears and the simplicity of operation of the system becomes evi- 
dent. 

Some semi-laboratory set-ups of equipment and the results ob- 
tained with them will next be considered. In Fig. 8 is shown an 
arrangement for IR System photography on a miniature set, using 
flashing gas-filled lamps. The motion picture camera is mounted on 
the small table to the left together with a monitoring finder arranged 
for IR System operation and a group of timers whereby the flashing 
of any regional lighting can be timed as desired. These units are 
synchronously driven and may be either combined or separated as 
desired. Suspended from the large central framework is a group of 
tubular lamps each of which, by the use of a suitable reflector, is 
arranged to illuminate appreciably only a single region. On the table 
under the framework, in the field of the camera, is a group of test- 
charts which can be readily moved to any desired position and which 
form a convenient subject for test photography. Mounted on the 
framework to the right is a group of power-supply units for the 
lamps, one power unit being provided for each region. These units 
are essentially high- voltage rectifiers charging condensers which, in 
turn, discharge through the gas lamps when these are suitably 
triggered. Simple methods of triggering the lamps without mechani- 
cal or electrical complications were devised. The timing of the lamps 
(that is, the triggering impulses) was derived from the regional timer 
synchronized with the camera operation as referred to above. The 
camera, as well as the finder, each contained a diffo. It was interest- 
ing in working with this equipment to be able to focus each region of 
of the set independently by shifting the timing of the corresponding 
regional illumination, this obviously being a startling and un- 
precedented procedure. 

In Fig. 9 is shown an enlargement of a single frame of 35-mm film. 
A 50-mm lens operating at approximately //2. 2 was used. The fore- 
ground toys were 5 feet from the lens; the background toys 15 feet. 
The toys were placed on a checkered fabric which also formed the 
background. The depth of the set corresponded to approximately 5 
regions. Fig. 9 illustrates the result obtained with standard photog- 
raphy. The toys in the second region are obviously not in sharp 
focus and the background is unidentifiable. 

In Fig. 10, on the other hand, are shown the results obtained by 
IR System photography using flashing gas lamps on identically the 



26 



ALFRED N. GOLDSMITH 



[J. s. M. P. E. 




FIG. 9. Enlargement of 35-mm frame with ordinary photography of minia- 
ture set. F 50-mm//2.2; foreground 5 feet, background 10 feet. 




FIG. 10. Enlargement of 35-mm frame with IR System of photography 
of miniature set. Same lens, aperture, film, development, and enlargement 
as for Fig. 9. 



Jan., 1942] 



THE IR SYSTEM 



27 



same set. Needless to say the same lens, aperture, film, development, 
and enlargement ratio were used in Figs. 9 and 10. 

Fig. 11 illustrates a small set illuminated according to the IR 
System by means of shuttered incandescent lamps. As a matter of 
convenience, these lamps were mounted on pipe racks. The shutter 
of each lamp was regionally synchronized and operated in synchro- 
nism and suitable phase relation to the camera and diffo drive. It 
was also possible to take pictures according to standard photography 
on this set-up either by removing the diffo from the camera or by 




FIG. 11. Shuttered incandescent-lamp illumination arrangement of set ac- 
cording to IR System. 

bringing all regional illuminations into the same timing (for example, 
that of Region 1). 

Using a lens of 2-inch focal length operated at //2, pictures were 
taken by standard photographic methods with four men between 
approximately 7 feet and 25 feet from the camera. The result is 
shown in Fig. 12. Only the foreground individual is in focus. The 
calendars in the middle ground should be noted, and are obviously 
badly out of focus. The poster behind the man in the background is 
indistinguishable. 



28 ALFRED N. GOLDSMITH [J. S. M. P. E. 

Fig. 13 shows the corresponding result when IR System operation 
was utilized. The same lamps, lens, aperture, film, development, en- 
largement, and all other pertinent circumstances apply to both Figs. 
12 and 13. The appearance of the calendars and poster should again 
be noted. 

A somewhat quantitative comparison between standard photog- 
raphy and IR System photography is shown in Figs. 14 and 15, 
respectively. Like Figs. 12 and 13, these are enlargements of single 
frames of 35-mm film. In each case the foreground plane was ap- 
proximately 11 feet in front of the right-hand calendar. The left- 
hand calendar was approximately 15 feet from the lens, and the back- 
ground poster at 25 feet from the lens. These pictures represent the 
empty set corresponding to Figs. 12 and 13, i. e., with the persons 
absent. Close microscopic examination failed to detect any difference 
between the sharpness of the test-charts in Fig. 15 at various distances. 

8. MULTIPLE SIMULTANEOUS ANGLE SHOTS 

Using systems wherein relatively brief flashes of timed illumination 
are employed in each part of the set, it is possible to photograph the 
same subject matter "simultaneously" from different angles. It is 
true that different portions of each frame period of l/24th of a sec- 
ond are used for the photography of each angle shot but the final 
films give no indication of this slight time staggering between the 
various angle shots. The method can be illustrated by consideration 
of Figs. 16 and 17. 

In the upper portion of Fig. 16 is shown one method of time 
division between two cameras. The periods during which the shutter 
of each camera is open, and the periods during which it is closed are 
indicated. Consider the period shown to the left during which the 
shutters are open. The length of this period may be taken as l/48th 
of a second on the arbitrary assumption that a 180-degree shutter is 
used. This period of l/48th of a second is equally divided by the 
vertical dot-dash line. The period of l/96th of a second to the left of 
the dividing line is assigned to Camera 1. Assuming 5-divisional 
operation, the flashes in the respective divisions occur as indicated by 
the vertical lines which are designated, respectively, F11-F15. The 
period of l/96th of a second to the right of the dividing line is as- 
signed to Camera 2 Assuming this also to be a 5-divisional take, the 
divisional light flashes will occur as indicated by the vertical lines 
F21-F25. So far as photography is concerned the arrangement thus 



Jan., 1942] 



THE IR SYSTEM 



29 




FIG. 12. Enlargement of 35-mm frame with ordinary photography of set. 
F 2 inches //2; men, respectively, at 7 feet 6 inches and approximately 11, 
15, and 25 feet. 




FIG. 13. Enlargement of 35-mm frame with IR System photography of set. 
Same distances, lamps, lens, aperture, film, development, and enlargement as 
in Fig. 12. 



30 



ALFRED N. GOLDSMITH 



[J. S. M. P. E. 




FIG. 14. Empty set corresponding to conditions of Fig. 12; ordinary 

photography. 



2 3 4 S S 
9 1011121314 
16 1? 18 IS 28 1\ 
2324252SHE8 

: 31 > 




FIG. 15. Empty set corresponding to conditions of Fig. 13; IR System 

photography. 



Jan., 1942] 



THE IR SYSTEM 



31 



far described is adequate and enables the simultaneous photography 
by the IR System of two different angle shots photographed, re- 
spectively, by Cameras 1 and 2. However, there would be only 24 
flashes per second on each portion of the set using this arrangement, 
and this would give rise to noticeable flicker. To minimize or eliminate 
this flicker, additional flashes API 1-A F15 andAF21-AF25 are intro- 
duced as shown during the period when the shutters of the two cameras 
are closed. While these additional flashes do not produce any photo- 













c 


/c 


f 


V 
















f? 


W 


/rr/ , 


/F 


,<! 
















< 


?/ 


* 


V 
















.? 


y/ 


/^7V 


-/f 


,1 












































C 


/OS 


f, 


9 






































< 


^^ 


Ci7A 


^ 














































































































































































































1 


















/ 






. 




(:' 


|/ 




- n 


' 







5 


o 




















































*f 


' 


1 


; 


f 




*f 


^ 


t 


if 




t 


fp, 








f 1 


^ 







i 























































\ s 



' t 1 



gwrree &/t/rrf& 

GlOSfO- CAM&e/t f OfN- C/>MGA / CiOSFG ~ C/9/rtf&A / OffM- C/>M/e/> 7 

O/^fAf - C/>M/e/>8 CLOSE O-CAM&e/t Of>N - G/IMfffA 2 CLOSED- C/>?e/> '3 



FIG. 16. Illumination time division for simultaneous multiple-angle photog- 
raphy in IR System. 

graphic effect they do minimize flicker since there are now 48 flashes 
per second illuminating each portion of the set. 

An alternative arrangement of somewhat different type is shown 
in the lower portion of Fig. 16. In this case the two cameras are so 
operated that when the shutter of one is open and the set is illumi- 
nated, the shutter of the other is closed and the set is not illuminated 
by its associated lights. The divisional illuminations for Camera 2 
are here indicated by F11-F15, while those for Camera 1 are shown 
as F21-F25. The avoidance of flicker in this arrangement requires 



32 



ALFRED N. GOLDSMITH 



[J. S. M. p. E. 



some planning in the placement and use of light, and for this reason 
the arrangement shown in the upper portion of Fig. 16 will sometimes 
be preferred. Experiments showed the feasibility of operation accord- 
ing to the methods of Fig. 16 whereby several cameras are each 
associated with their timed illumination of the set, and with the 
divisional light distribution utilized in the IR System. 

A schematic layout for the lights corresponding to the arrangements 
of Fig. 16 is shown in Fig. 17. The three-walled set is indicated. 
Camera 1 takes a comparatively wide-angle shot wherein the person 
C, who is the center of interest, is photographed in a semi-close-up. 



CAMERA 







FIG. 17. Illumination and camera arrangements for simultaneous multiple 
angle shots according to IR System. 



The illumination is produced by the divisional lamps L11-L14. 
Since this is to be a semi-close-up, the illumination may be kept com- 
paratively soft. The lamps are coordinated with the camera and 
diffo as indicated by the dashed lines marked Co-ordination 1 . Camera 
2 takes a longer shot but with a narrower angle of a different section of 
the set as well as a partial side view of the individual C. Since this is 
a more distant view, the lighting may be made harder to emphasize 
detail and modelling. The corresponding lamps are L21-L24. In 
this case also, the lamps are timed in association with the camera and 
diffo as indicated by the dashed lines marked Co-ordination 2. It 
should be particularly noted that entirely different lighting effects 



Jan., 1942] THE IR SYSTEM 33 

can be produced in the pictures taken simultaneously by the two 
cameras a markedly novel and obviously useful feature. 

9. STUDIO LIGHTING METHODS FOR INCREASED-RANGE SYSTEM 

In essence, conventional studio lighting may be arbitrarily divided 
into key lighting and modelling lighting. The former is a more or less 
general illumination, intended to set a minimal lighting level through 
all or most of the set. The latter is lighting which is intended to 
produce shading, thus giving roundness, form, or solidity to persons 
or objects. It is generally carefully directed and of the spot variety. 

In the IR System, modelling lighting requires relatively little 
change from the previously used conventional methods. Some care is 
taken to use a sufficient number of aimed spots to avoid undue space 
spillover while yet retaining the desired modelling for rather wide 
axial excursions of the actors. In the frontal regions this requires 
reasonable care. In the rear regions the latitude in light placement 
is so wide as to eliminate any necessity for special treatment. 

Key lighting by the IR System requires suitable methods for 
duplicating the lighting vector in any important portion of the set. 
That is, the intensity and direction of the incident light must be 
duplicated in each region, and without undue space spillover of light 
from that region. It has been found in practice that any desired 
light vector normally encountered in motion picture photography 
can be duplicated by IR System lighting after a brief analysis of 
what is required. As experience in the use of the IR System is 
acquired, this duplication becomes increasingly simple. The details 
of light placement and the securing of specific effects using IR System 
lighting can not be here included. To some extent, as with all other 
lighting, it involves solving minor lighting problems as one goes along. 

10. PRESENT OPERATING CONSIDERATIONS 

The following comments list certain minor limitations in the IR 
System which, in the main, are only apparent. Methods of overcom- 
ing these seeming restrictions are known, except where otherwise 
stated. 

(a) Need for Controllable Lighting. Since the regional or divisional 
control of lighting is an essential feature of the IR System it is 
obvious that the system can be best used under studio conditions. 
The evolution of modern light-sources makes their control practicable. 

(b) Registration Effects. The magnitude of the registration 



34 ALFRED N. GOLDSMITH [J. S. M. p. E. 

effects is controllable, and all indications are that such effects can be 
made pictorially imperceptible or negligible. A number of compara- 
tively simple equipment designs and operational methods have been 
devised for this purpose. 

(c) Lighting Efficiency. The lighting efficiency of the IR System 
depends upon the type of lamp which is used. Flashing lamps, wherein 
all radiated light is emitted during an extremely brief period, enable 
efficient operation. If arcs or incandescent lamps are used, with 
suitable shutter arrangements, the total amount of light produced by 
each lamp must be correspondingly greater than that which passes 
through the shutter opening to the related region or division of the set. 
The shutter naturally prevents the utilization of the light not passing 
through its opening. Thus, for shuttered lamps, while the amount of 
light on the actors does not exceed that used in ordinary photography, 
the total amount of light flux produced by the lamp (and consequently 
the power load and the studio air-conditioning load) must be in- 
creased. On the other hand, when flashing lamps (e. g., gas lamps) 
are used, and provided these have the high efficiency of the better 
designs of this type of lamp, the light on the actors and also the total 
light flux produced by the lamps are each no greater than for usual 
photography. Thus the flashing lamp, although the more unusual, 
is in some respects more convenient and efficient. 

(d) Flicker. Inasmuch as the illumination is intermittent, pre- 
cautions are taken to prevent visible or objectionable flicker. In 
Fig. 16, a 48-flash-per-second regime was illustrated. However, the 
lighting frequency on each portion of the set can be increased as de- 
sired. Experience has shown that 48-cycle illumination, while not 
absolutely smooth, does not seem to create sufficient flicker to be 
annoying to most people. 

(e) Acoustic Interference. Shutters, driving motors, or any other 
equipment for intermittent illumination timing must be silenced to an 
extent such that no sound from them is recorded. The expedients 
adopted for this purpose are in considerable measure conventional. 

(/) Inductive Interference. Flashing lamps, and their associated 
timing circuits, require suitable electrical design and shielding to 
avoid any undesired induction, for example, into the recording circuits. 

(g) Approach Shots. The maintenance of sharp focus in all parts 
of the set throughout approach shots 13 readily possible in the IR 
System. It requires the use of one of a number of expedients mainly 
based on a consideration of appropriate regional and divisional ar- 



Jan., 1942] THE IR SYSTEM 35 

rangement of the lighting. The detailed methods can not be included 
within the space of this paper. 

(ti) Panoramic Shots. It is also possible readily to make pano- 
ramic shots using the IR System, the lighting arrangements suitable 
for the purpose being known and practicable. Experience and analy- 
sis have shown that the adaptation of regional illumination to a wide 
variety of photographic procedures is simpler than might be antici- 
pated. 

11. FUTURE DEVELOPMENTS 

While the IR System is inherently an optical development, it is 
believed that it will have profound effects on the dramatic and pictorial 
results to be obtained in the fields of still-picture advertising and the 
like, on entertainment and educational motion pictures, on television, 
and on a number of related fields. Dramatically, it should enable a 
gradual blending of the arts of the legitimate theater and the motion 
picture studio, as a result of which the motion picture will gain added 
continuity, fluidity, smoothness, and naturalness, all of which should 
add to its appeal. Further, the dramatic freedom of the actor will be 
enhanced, and the reduction in shooting time will reduce the strain on 
him. The matching of acting in all parts of a prolonged sequence 
will be more nearly automatic and perfect, as will be the matching of 
the sound with the picture in all parts of a take. 

12. CONCLUSION 

In relation to the work described in this paper it is a pleasure to 
acknowledge the capable cooperation of certain of my co-inventors 
and associates. In particular, I am indebted to Mr. Harry R. 
Menefee for the many original and useful thoughts as well as practical 
constructions which he has contributed, to Mr. Fritz Kastilan for the 
elaborate optical calculations and the careful designs which he 
originated, and to Messrs. Bernon Woodle and Remsen Donald for 
their painstaking calculations and laboratory activities during the 
progress of the work. And I am especially indebted to Mr. Court- 
land Smith for the inspiration, thoughtful executive planning, and 
courage which he has constantly displayed and which have made 
possible the prosecution of this project as well as its successful out- 
come. 



A NEW DICHROIC REFLECTOR AND ITS APPLICATION 
TO PROTOCELL MONITORING SYSTEMS* 



G. L. DIMMICK** 



Summary. Certain crystals have long been known to transmit light of one 
color and reflect light of another color. Some thin metallic films also exhibit the same 
phenomena. By evaporating alternate layers of high and low-index insulators on 
glass, it is possible to produce a surface having predetermined transmission and re- 
flection characteristics, and having no appreciable absorption. A method of determin- 
ing the properties of multilayer films is described. 

Design details and curves are given for a three-layer film having its peak transmis- 
sion at 4400 A and its peak reflection at 7920 A . This film is successfully employed 
in high-level photocell monitoring systems for sound recorders. Nearly all the actinic 
value of the modulated light is transmitted to the photographic film while a large part 
of the red and infrared is reflected to a cesium photocell for monitoring. 

It has long been known that thin films of some metals are selective 
in their ability to reflect and transmit light. A thin film of gold is 
quite transparent to green light and shows strong selective reflection 
for the red and yellow region. Many aniline dyes appear to have 
one color when viewed by reflected light and another color when 
viewed by transmitted light. The material possesses what is known 
as a surface color, and the transmitted light gets its color by being 
deprived of certain rays by reflection at the surface and certain 
others by absorption in the interior. 

There is another type of selective reflector which depends upon the 
interference of light in thin films. This type is far more efficient be- 
cause the absorption is usually negligible. In its simplest form this 
reflector consists of a single thin film between two transparent media. 
A soap bubble and a layer of oil on water are perhaps the most com- 
monly experienced examples of this type. 

If we wished to make use of the interference principle to obtain a 
selective reflector capable of reflecting a large percentage of light in a 
narrow region of the spectrum, we should find that the single thin 

* Presented at the 1941 Fall Meeting at New York, N. Y.; received October 
20, 1941. 

** RCA Manufacturing Co., Indianapolis, Ind. 

36 



A NEW DICHROIC REFLECTOR 37 

film would be inadequate for the purpose. Both the intensity and 
the purity of the reflected light may be increased through the use of 
multiple films arranged in alternate layers having different indices 
of refraction. As is often the case nature has already accomplished 
this result and with a remarkable degree of precision. Crystals of 
chlorate of potash are sometimes found to reflect very brilliant and 
pure colors. In 1852 Sir George Stokes investigated this phenomenon 
and found it to be due to the existence of planes within the crystal at 
which periodic "twinning" had occurred. Later, in 1888 Lord Ray- 
leigh studied these crystals and found that the spectrum of the re- 
flected light consisted of a narrow band only about twice the width 
of the D lines. From this he calculated that a crystal plate 0.14 mm 
in thickness contained about 700 reflecting planes. 

Thin films of many metals and dielectrics may be produced by 
heating the materials in a vacuum and allowing the vapor to deposit 
upon a plate of glass. This very important process lias been known 
for a long time, but it is only in the last few years that it has received 
the attention which it deserves. Dr. J. Strong did important work 1 
in applying the evaporation method to the production of aluminum 
films for mirrors. Later he evaporated thin films of calcium fluoride 
on glass and observed a considerable reduction in the surface reflec- 
tion. In 1939 Dr. C. H. Cartwright worked on the problem at the 
Massachusetts Institute of Technology. He not only improved the 
process of reducing reflection from glass but also produced selective 
reflectors by evaporating multiple films on glass. 2 The author's 
interest in the subject began when he visited Dr. Cartwright and saw 
some of his work. 

A rigorous mathematical analysis of multiple films would be quite 
involved and would not be of any help in giving us a physical picture 
of what happens within the films to produce the final result. The 
graphical method of analysis is far more flexible and is the one that 
the author has made use of in the design of a selective reflector having 
predetermined characteristics. Although the method described be- 
low is not strictly accurate, it yields a result which is consistent with 
our meager knowledge of the optical constants of evaporated ma- 
terials. 

Fig. l(a) shows a film having a refractive index of 1.222 on the sur- 
face of a glass plate having a refractive index of 1.500. A ray of 
light AQ is passed through the film from the glass plate into the air. 
The primary reflections from the two surfaces are shown as A\ and 



38 



G. L. DlMMICK 



[J. S. M. P. E. 



A 2 . The amplitudes of the two reflected rays are determined from 
the formula 




and 



ffi-1 

, N* 



These expressions for the amplitudes of the reflected rays are 
quite accurate when the angle of incidence of A o is less than 30 de- 
grees. 

Fig. 1 (b) shows the graphical method of determining the resultant 
amplitude A r of the two reflections. The lengths of the vectors A\ 
and Az are laid off in proportion to the calculated values. The vec- 
tor A i is stationary and positive in sign. 




Ar 



(b) 

FIG. 1. (a) shows a single low index film on glass, (b) shows the arrange- 
ment of vectors for the film in Fig. l(a). 

The vector Az coincides with A\ when the film thickness is zero, 
and rotates in a counterclockwise direction as the film thickness in- 
creases. The angle between the vectors is determined from the 
equation 



where X is the wavelength of the light and t is the film thickness. Thus 
when N\t = X/4, then B = TT radians, or 180 degrees. The vectors 
are opposed in direction and equal in amplitude, making the resultant 
reflection zero for a particular wavelength of light. If Nit has other 
values than X/4 or multiples of it, the resultant amplitude A r is de- 
termined graphically by the parallelogram method and the reflec- 
tivity is obtained from the relation 



Jan., 1942] 



A NEW DICHROIC REFLECTOR 



39 



*--:)' 



where AQ is the amplitude of the incident light. 

Suppose the film is made to have an optical thickness of A/4 at 
5100 A and our problem is to determine the curve of reflection vs. 
wavelength between the limits of 4000 and 7000 A. Fig. 2 (a) shows 
the angular position of vector A% for different wavelengths. The 
angle between A\ and A^ for any wavelength is given by the expres- 
sion 



5100 



X 180 





4000 



7000 



(a) 



in 
(b) 



FIG. 2. (a) shows the arrangement of vectors for different wavelengths. 
(6) shows the relationship between per cent reflectivity and wavelength for 
the single film in Fig. l(a). 

Thus for 7000 A the angle is 131 degrees and for 4000 A the angle 
is 229 degrees. For each wavelength the vectors A\ and A% are added 
graphically and the resultant amplitude A r is squared to obtain the 
reflected intensity. Fig. 2(b) shows how the reflectivity varies with 
wavelength when the optical thickness of the film is X/4 at 5100 A. 

In arriving at this curve it will be observed that no account was 
taken of the change in index of refraction with wavelength for the 
glass plate or the film. If we knew the index of refraction at dif- 
ferent wavelengths for both these materials, it would be quite easy 
to take this factor into account. In Fig. 3 the vectors A* and A\ 
would have different amplitudes for each wavelength instead of hav- 
ing a constant amplitude as shown. For each wavelength the am- 
plitude would be determined from the relations of equations 1. If 
the dispersion is no greater than the value for ordinary crown glass, 



40 G. L. DlMMICK [J. S. M. P. E. 

our maximum error would not be more than 6 per cent if we disre- 
garded it. 

In addition to the primary reflected ray Az in Fig. l(a), there ex- 
ists secondary and tertiary rays A '2 and A " 2 , shown as dotted lines. 
The effect of these rays also can be included in the graphical method. 
The vector A' 2 would be drawn one-hundredth as long as A 2 and its 
angle of rotation would always be made just twice that of A 2 . The 
vector A "2 would be one ten-thousandth as long as A 2 and its angle of 
rotation would always be made just three times that of A 2 . All the 
vectors would be added graphically in order to determine the resul- 
tant amplitude under any given condition. It can be seen that if the 
individual primary reflections are relatively small, the effect of sec- 
ondary and tertiary reflections is negligible. 

The analysis of the single film on glass has been made in order to 
illustrate the ease with which the properties of thin films can be de- 
termined by the graphical method. We shall now make use of the 
method to design a multifilm selective reflector to meet a given set of 
practical conditions. When completed, the reflector is to be placed 
in the light path of a sound-recording optical system between the slit 
and the objective lens. Its purpose is to transmit most of the actinic 
rays of light required to expose positive film and to reflect a large 
part of the red and infrared light to a cesium photocell for monitoring 
purposes. The advantages of such a mirror are obvious. It per- 
mits us to have a high-level monitoring system relatively free from 
hiss, microphonic disturbances, and noise due to electrostatic or 
magnetic pick-up. It maintains a rigid check upon the quality of 
the modulated light reaching the film, without appreciable reduction 
in exposure. 

Ordinary positive film receives its greatest exposure from an in- 
candescent lamp at about 4400 A. A cesium photocell receives its 
greatest energy from an incandescent lamp at about 7500 A. Our 
requirements for the selective reflector are that its peak transmission 
occur at 4400 A and its peak reflection occur at 7500 A. We shall 
try a three-layer film to see whether it meets these requirements. 
Fig. 3 (a) shows a plate of glass with two high-index films and one 
low-index film upon its surface. 

The deciding factor in choosing values for N 2 , N 3 , and N* (Fig. 3a) 
was that materials are available that have these indices of refraction. 
In order to obtain the greatest peak transmission at one wavelength 
and the greatest peak reflection at another wavelength, it is necessary 



Jan., 1942] 



A NEW DICHROIC REFLECTOR 



41 



so to arrange the films that the algebraic sum of the four reflected 
amplitudes is as great as possible and that the algebraic sum of one 
group of amplitudes is equal to the sum of the remaining amplitudes. 
The first condition is obtained by placing the high-index film on the 
outside in contact with the air. When the glass plate is made to 
have an index of refraction of 1.545, the second condition is also met. 
In other words, A\ + A% + A z + ^4 is greater than for any other 
possible arrangement of the three films, and A\ + A = A 2 + A 9 . 



A= loo 




A, Ay 



(b) 



FIG. 3. (a) shows a 3-layer film on glass. (6) shows the arrangement of 
vectors for zero film thickness, (c) shows the required arrangement of 
vectors at 4400 angstroms. 



The next problem is to determine the proper thicknesses for the 
three films. Fig. 3(6) shows the arrangement of the vectors when 
the thickness of all films approaches zero. The vectors resulting 
from light reflected in passing from a medium of higher to a medium of 
lower index are given one sign. The vectors resulting from light re- 
flected in passing from a medium of lower to a medium of higher 
index are given the opposite sign. The reason for this is that a phase 
reversal takes place when light passes from a dense to a rare medium 
but there is no phase change when it passes from a rare to a dense 
medium. The author prefers always to give the stationary vector 
AI a positive sign and let the above considerations determine the 
signs of the other vectors. 



42 



G. L. DlMMICK 



[J. S. M. P. E. 



Fig. 3(c) shows the necessary arrangement of the vectors at 4400 
A if the transmission is to be maximum. Comparison of Figs. 3(6) 
and 3(c) shows that vector A 2 has been rotated through 2ir radians, 
or an even multiple of this angle. Since the optical thickness Ntfi = 
0X/47T, it follows that the minimum thickness of the first film is X/2. 
Again comparing Fig. 3(b) and Fig. 3(c), it is evident that vector A 3 
has been rotated through an odd multiple of TT radians. But the 
first film has already caused A 3 to rotate through 2?r radians. The 
first odd multiple will then be 3?r radians, and this makes the optical 
thickness of the first two layers 3X/4 and the optical thickness of the 



e 
"40 




<o 

FIG. 4. (a) shows the arrangement of vectors at 4400 angstroms. 
(6) shows the arrangement of vectors at 7340 angstroms, (c) shows the re- 
lationship between per cent reflectivity and wavelength for the selective 
reflector. 

second layer X/4. A final comparison of Fig. 3(6) and Fig. 3(c) 
shows that vector A* has been rotated through an odd multiple of 
TT radians. The first two films have already rotated A* through Sir 
radians, so the first odd multiple will be 5?r radians. This corre- 
sponds to an optical thickness of 5X/4 for all three films or X/2 for the 
third film. This completes the design of the three-layer selective 
reflector. 

We now have to find out whether this reflector meets our require- 
ments. To do so we lay out the positions of the four vectors for 
several different wavelengths and find the resultant amplitude in 
each case. The square of the resultant amplitude gives us the re- 
flected intensity for the wavelength in question. 



Jan., 1942] A NEW DlCHROIC REFLECTOR 43 

Fig. 4 (a) shows the position of the four vectors for a wavelength 
of 4400 A and Fig. 4(6) shows their position for a wavelength of 7340 
A. The angular positions 2 , 8s, and 4 of the vectors A 2 , A 3, and A 
(Fig. 4&) are determined for any wavelength by the formulas 

* = 4400 x 360 . ft = 00 x 540 , = MpO x 900<1 

A X A 

where X is the wavelength in question and 6 is the angle of a particular 
vector as measured from its position shown in Fig. 3(b). This is 
the position when the film thicknesses approach zero, but it also rep- 
resents the position when the wavelength approaches infinity. 




FIG. 5. Shows the position of the reflector in a standard 
recording optical system. 

Fig. 4(c) shows a curve of reflectivity vs. wavelength for the three- 
layer selective reflector (Fig. 3a) as determined by the graphical 
method. According to this curve the transmission of the reflector is 
100 per cent at 4400 A and the reflectivity is 84 per cent at 7340 A. 
Because of the assumptions that have been made, it is not to be ex- 
pected that the actual reflector will reach values quite as high as in- 
dicated above. Measurements were taken on many reflectors made 
to the above specifications. These showed that the transmission is 
95 per cent, when measured with a filter and photocell having the 
same color characteristics as positive film. The reflectivity as mea- 
sured through a Wratten A filter is 65 per cent. In view of the rather 
broad band width of these filters, it is felt that the measurements rep- 
resent a reasonably good confirmation of the theory. 

Photocell monitoring systems utilizing the new selective reflectors 
have been in use in two Hollywood studios for several months. The 
reflectors were inserted into the objective barrels of standard van- 



44 G. L. DIMMICK 

able-density and variable-area optical systems as shown in Fig. 5. 
The three-layer film was applied to the bottom surface of a glass 
wedge 14 which was placed at a 45-degree angle in the optical path 
between the slit 11 and the objective lens 12 and 13. A photocell 17 
was placed directly beneath the reflector. Several lenses (not shown) 
between the reflector and the photocell serve to direct the light from the 
slit upon the two push-pull cathodes. The angle and the thickness of 
the wedge 14 were such as to cause no displacement of the light-beam at 
the back of the objective 12 when the reflector is inserted in a standard 
optical system. The angle of the wedge is just sufficient to cause the 
light 18 and 19, reflected from the upper surface 16, to be thrown off 
the edge of the objective lens and the photocell. This results in the 
practical equivalent of a single-surface reflector, and completely elimi- 
nates the double image of the slit that would result if the two faces 
15 and 16 were parallel. 

REFERENCES 

1 STRONG, J. : "On a Method of Decreasing the Reflection from Non-Metallic 
Substances," /. Opt. Soc. Amer., 2b (Jan., 1936), p. 73. 

2 CARTWRIGHT, C. H., AND TURNER, A. F. : "Multiple Films of High Reflecting 
Power," Phys. Rev., 55 (June, 1939), p. 595. 



PRODUCTION AND RELEASE APPLICATIONS OF FINE- 
GRAIN FILMS FOR VARIABLE-DENSITY 
SOUND RECORDING* 

C. R. DAILY AND I. M. CHAMBERS** 



Summary. Fine-grain film materials have supplanted the normal positive type 
emulsions for all variable-density sound-recording and printing operations. The 
sound-quality improvement realized by the reduction in noise and distortion is now 
available for all sound operations, including release prints. The paper describes 
a number of problems encountered and solved in the commercial application of such 
films for sound recording, including factors affecting the choice of negative and print 
materials, noise, distortion, sensitometric characteristics, recorder lamp supplies, 
and noise problems on stages. 

A complete fine-grain sound-recording program is now in operation 
at this Studio for variable-density sound-track, a fine-grain negative 
stock being used for all production and release and a fine-grain posi- 
tive stock tor all dubbing prints and black-and-white release prints. 
Daily sound prints are also made on the out-take fine-grain negative 
stock. This improved film-recording and release program has been 
in operation since March, 1941, and represents the culmination of 
over three years of work on the general problem of adapting fine- 
grain films for sound recording and release picture and sound use. 
Over two years ago an earlier type of fine-grain film was first used on 
a large scale for sound dubbing prints and later the same film was 
used for a limited amount of release negative and release printing, 1 
while now the improved films are used for all operations. While a 
substantial improvement in sound quality has been the major objec- 
tive of this project, a gratifying benefit to the release picture quality 
has also been obtained. The presentation of orchestral and vocal 
music has been materially improved in clarity and definition due to 
the increased volume range made available. "Swish" is no longer a 
major source of trouble, permitting piano, guitar, drum, and similar 

* Presented at the 1941 Fall Meeting at New York, N. Y. ; received October 
15, 1941. 

** Paramount Pictures, Inc., Hollywood, Calif. 

45 



46 C. R. DAILY AND I. M. CHAMBERS tf. S. M. P. E. 

types of percussion instrumentation to be presented with improved 
fidelity. Dialog quality has also been improved, voices standing 
out with better screen presence, an important factor in release to the 
theaters. Film distortion has been reduced due to (a) improved 
methods of development, (b) the characteristics of the film stocks 
themselves, and (c) to the use of ultraviolet light for printing. Proc- 
essing latitudes are also wider than formerly obtained and the high- 
frequency response has been increased due to the improved resolu- 
tion of the film. 

Some of the problems encountered and solved in connection with 
the commercial application of these new films will be described in this 
paper. Messrs. Wilkinson and Eich 2 describe the laboratory modifi- 
cations and procedures in connection with the development and 
printing of such films for release. 

At the Fall Convention of the Society in October, 1939, a paper 1 
was presented describing methods employed at that time in connec- 
tion with the first commercial release application of fine-grain film 
stocks. The experience gained in connection with these first releases 
had clearly indicated that a marked reduction in film noise would 
be accompanied by an improvement in sound quality, and the re- 
cent large-scale use of the new fine-grain films has completely verified 
this earlier observation. 

The fine-grain negative and print stocks currently used for sound 
recording represent a definite improvement over the types of fine- 
grain films available two years ago. Tids improvement has been 
obtained through the testing of a series of experimental film stocks, 
to determine their adaptability to the numerous production require- 
ments. Since the desired characteristics of the sound negative stock 
are dependent to some extent upon the characteristics of the release 
positive to be used, it was first necessary to obtain a fine-grain re- 
lease positive stock which would satisfy picture requirements. 
When such a satisfactory stock for picture release became available 
during the latter part of 1940 the Laboratory undertook the modifica- 
tion of the daily and release printers to provide the necessary increase 
in light. Since it was also known that sound quality would be further 
improved by the use of ultraviolet light in the printing to the fine-grain 
stock, sufficient light-intensity was made available in the printer 
modification to permit sound printing with a 2-mm Corning 584 
filter. This fine-grain release positive material makes possible a 
static reduction of 6 to 8 db in film noise and a swish reduction of 10 



Jan., 1942] FlNE-GRAIN FlLMS FOR SOUND RECORDING 47 

to 15 db in the release to theaters compared with the former standard 
release print stocks. While a still finer grained stock would be 
desirable, the benefits obtained from this stock are definitely worth 
while and valuable experience is being gained in the handling of such 
materials. 

At approximately the same time that the fine-grain release positive 
material became available, a suitable fine-grain negative also was 
developed which could be printed to the release positive material 
with a further improvement in sound quality, and with an added in- 
crease in signal-to-noise ratio because of the added reduction in nega- 
tive noise. This new negative material is essentially as fine-grained 
as the new release positive and also possesses satisfactory develop- 
ment characteristics when exposed with tungsten light on the existing 
production recorders. 

Tungsten recording was considered necessary for the production 
recorders because of the added weight and power-supply problems 
which would have to be met if arc sources were used, particularly 
for location work. It was found possible to obtain sufficient density 
en the new negative stock with a tungsten lamp by taking advantage 
of (a) high-gamma negative development for use with ultraviolet 
printing, (&) a modification of the negative developer to give the 
highest density consistent with stable development practice and low 
distortion, and (c) by coating the lens surfaces of the recorder optical 
system. 

The new negative stock is also suitable for pre-selection sound 
printing, an important requirement in connection with the negative 
pre-selection practice employed. All sound negative takes are broken 
down into two groups at the end of each day's work. The "O.K." 
takes that are to be printed are spliced together and sent to the Lab- 
oratory for immediate development, while the exposed "out" takes and 
non-exposed short ends are spliced together in a second group and 
held undeveloped. These "out" takes are identified so that any one 
of them can be readily located and developed at a later date if needed 
by the Production Department. Since takes may have to be de- 
veloped days or weeks after they are recorded, it is desirable that 
minimum changes in characteristics be experienced during this time, 
such as loss of density and change of gamma. The new negative 
stock is quite satisfactory in this respect, based on a six-month use. 

When a production is completed, the "out" takes are sent to the 
Laboratory to be used for daily sound print stock. Since the new 



48 C. R. DAILY AND I. M. CHAMBERS (J. S. M. p. E. 

negative stock is essentially as fine grained as the fine-grain positive, 
the daily prints are very quiet and their quality is comparable to the 
dubbing prints made on the fine-grain positive. The same positive 
developer and the same type of ultraviolet printing light is used as 
for release printing, with only a slight difference in exposure. 

The splicing of the takes of negative stock for either negative de- 
velopment or for subsequent daily print use is accomplished satis- 
factorily by scraping both sides of the film, cementing with normal 
Roscoe cement and allowing a 15-second drying time. The splices 
in the takes to be developed as a negative are further reinforced by 
using Eastman film splicing tape, as it is very important to prevent 
these splices from pulling apart in the negative developer. Since 
the current splicing and reinforcing technic was adopted over four 
months ago, there has been no breakage of film in the negative de- 
veloper. 

A transition period of a few weeks' duration was necessary to con- 
vert the plant completely from the use of standard negative to fine- 
grain negative, involving changes in both the printing and negative 
development procedure. Since the fine-grain negatives required 
ultraviolet prints while the standard negatives formerly used re- 
quired white-light prints, it would have been impracticable to use 
both types of negatives on a given production without unduly com- 
plicating the dubbing print operation which is normally made from 
assembled reels of production negative. Therefore, when the plant 
was ready to start recording fine-grain negatives, new productions 
were started on the new stock while productions which were then 
being recorded on standard negative were continued in this manner 
to completion. The new negative stock could be developed in the 
same tank with the then current standard negative stock, effecting 
a definite economy in film handling during the transition period. 

Both the new fine-grain materials are quite hard and appear to be 
less susceptible to dirt pick-up than the older types of emulsions. 
This relative freedom from dirt and noise trouble is quite essential 
with fine-grain films because the reduction in film noise uncovers 
dirt noises that formerly would have been less important because of 
the masking effects of the film noise itself. While additional care 
is taken, the negatives stand up well during the cutting, cleaning, 
and printing operations; the positive material is giving satisfactory 
service in release. 

In order to increase the light-transmission efficiency of the produc- 



Jan., 1942] FlNE-GRAIN FILMS FOR SOUND RECORDING 



49 



tion sound recorders, their condenser and objective lenses were given 
a special surface treatment to reduce the amount of reflected light. 3 
This coating, which was applied to give the highest transmission at 
4400 A, increased the efficiency of the recorder optical systems by 
approximately 50 per cent permitting the 9-ampere, 10- volt lamps to 
be operated at 8.5 instead of 9.0 amperes. Thus a safe margin of 
exposure was available to take care 
of variations in lamp and stock 
sensitivity. 

The release recorders are equipped 
with an auxiliary optical system to 
permit squeeze-track recording, and 
require approximately 40 per cent 
more light than the production re- 
corders. In order to provide a safe 
margin of increased light-intensity 
on these machines, the tungsten 
lamps were replaced with air-cooled 
AH-8 high-pressure mercury arcs, 
in addition to coating the lenses. 
Since satisfactory operation of such 
arcs has been obtained with the 
temporary arc installation used two 
years ago, 1 the power and air-sup- 
ply systems for the arcs were com- 
pletely redesigned further to im- 
prove their operation. 

The arc itself was remounted in a 
new holder to permit rapid align- 
ment and to reduce any noise which 
might be caused by too turbulent an air-stream over the arc. The 
capillary and support- wire structure is remounted in a new socket 
shown in Fig. 1. An unmodified AH-8 arc also is shown for com- 
parison. To remount the arc, the envelope is removed, the support 
wires then being held in a jig while the glass press is removed. The 
lead wires are then straightened and inserted in suitable holes in the 
two connecting rods inserted in the metal-covered linen bakelite 
socket, these rods also forming the plug for the power leads. The arc 
and socket are then installed in a new type recorder lamp mounting 
shown in Fig. 2, which provides three linear and three rotational 




FIG. 1. High-pressure AH-8 
mercury arcs. (Left) modified arc, 
mounted in special socket; (right) 
unmodified arc. 



50 C. R. DAILY AND I. M. CHAMBERS (J. S. M. P. E. 

degrees of movement, permitting precision alignment of the arc on 
the recorder. Air is drawn through a glass-wool dust filter at one 
end of the housing, passes over the arc capillary, and thence through 
the exit tube to a rubber hose which connects to the exhaust fan 
mounted on the recorder base. A filter glass in the top of a sheet- 
metal housing, not shown, permits the operator to inspect the arc 
while in operation. 




FIG. 2. Arc mounting on film recorder. Contact pins appear 
at left; air inlet holes and glass-wool filter at left of glass; arc in 
glass tube; air exit at right; high-tension starting wire enters 
through bottom. 



The general design of the arc rectifier and self -regulating air-suc- 
tion system was adapted from a Metro-Goldwyn -Mayer design. 4 
Fig. 3 is a schematic diagram of the power-supply for the arc. The 
upper portion of this diagram shows a thyratron rectifying and filter- 
ing circuit with output voltage control obtained by phase adjustment 
of the voltage to the grids. This circuit, while designed for a maxi- 
mum load current of 3 amperes, is currently being used to provide 
0.7 to 1.1 amperes at 140 volts, the excess capacity being available for 
possible future use of light-filters or other improvements which might 
require more light. 



Jan., 1942] FlNE-GRAIN FlLMS FOR SOUND RECORDING 



51 



Without further regulation, the rectifier portion of this circuit per- 
mits an 18-per cent change in arc wattage to occur for a 10-per cent 
change in line voltage. Therefore, the regulating circuit shown in 
the lower portion of Fig. 3 was developed, which permits only a 3-per 
cent arc wattage variation for a 10-per cent line voltage fluctuation. 
The regulator applies a variable d-c grid bias to the thyratrons, 
controlled by variations in the a-c line voltage. This varying d-c 
voltage for the grids is obtained from the drop across a resistor in 
series with a VR-150-30 tube which is supplied in turn with recti- 
fied 180 volts from the a-c source. With a =* 10-per cent line- 
voltage variation, the drop across this resistor will vary between 




FIG. 3. Mercury arc power supply and. air-cooling system. 



12 and 48 volts and approximately 1 /i of this voltage is applied to 
the grid circuit in the proper polarity, to compensate for the line 
variations. 

The arcs are started each time they are to be used by externally 
exciting them with a high-frequency spark which is brought near the 
glass envelope by the high-tension cable shown in Fig. 2. Between 
takes the lamps are kept burning on a "hold" position, which keeps 
them hot at a reduced wattage. One hundred to 160 watts are 
needed to obtain the unbiased density of 0.65 currently used for the 
dubbed negative. The power-supply system maintains the desired 
wattage within =^3 watts, providing a very stable value of negative 
density. 



52 



C. R. DAILY AND I. M. CHAMBERS [J. S. M. P. E. 



With conventional types of sound-negative materials formerly 
employed it was necessary to introduce into the developer a certain 
quantity of bromide for a given density variation per quantity of 
film developed. After the transition had been made to the fine-grain 
stock it was found that this bromide concentration could be dropped 
to about one-third the former value for an equivalent density variation 
this reduction being made possible because the bromide released dur- 
ing development by the fine-grain stock is very much less than the 
amount released by the conventional negative materials. This re- 
duction in bromide concentration is therefore responsible for a con- 
siderable gain in negative density for a given gamma and exposure. 

































^"^ -tL 


*At*4 N&#rivt_ 




^X 


'^^^ 


















ITB-^ 


^ 


x 




1 










S0% V 


OOUtfilH 


*~i 


f 


L/0<" 


K//i^f 




L 












.^ 




1 








r 






x 


^x 


*? 


\ 















"- 


* 




















FIG. 4. II& and light-valve gamma character- 



istic of the fine-grain negative. 

II& and light-valve gamma characteristics for the fine-grain nega- 
tive stock are shown in Fig. 4. The 116 characteristic is exposed 
with tungsten light and the light-valve characteristic is typical of 
either tungsten or mercury arc exposure. A lib gamma of 0.72 and 
a light-valve gamma of 0.60 are shown for tungsten original recording 
with an unbiased negative density of 0.55, the negative gamma, how- 
ever, being adjusted as needed for various emulsions. While the 
lib gamma is higher than the light-valve gamma for this emulsion, 
it is a satisfactory index for laboratory control with a given developer, 
this observation being borne out by routine intermodulation tests. 

The projected printed-through characteristics of these negative 
characteristics to the fine-grain positive are shown in Fig. 5. The 
prints were made with an air-cooled AH-8 mercury arc using a 2-mm 
Corning 584 ultraviolet filter. 

In Fig. 6 is shown the projected print transmission vs. light-valve 
spacing obtained from a replot of the data shown in Fig. 5. This 



Jan., 1942] FlNE -GRAIN FILMS FOR SOUND RECORDING 



53 



static characteristic is essentially a straight line, which sensito- 
metrically confirms the low distortions measured dynamically with 
either single-frequency or intermodulation recordings. 

In Fig. 7 is shown a typical 60/400-cycle intermodulation distor- 
tion characteristic for this fine-grain film combination as a function 
of print density for typical original tungsten recording conditions of 
negative exposure and development. The minimum distortions 
shown are lower than those obtained with the conventional types of 
positive materials which were formerly employed, an observation 
which is confirmed by aural checks. The numerical magnitude of 



Q 



^ 



as/tf 



e r 90J^. 



X 



- 1 



tv* 



1.4 
LOG 



FIG. 5. Projected print density vs. log negative 
exposure for 116 and light-valve gammas. 

the minimum distortion as measured is a function to some degree 
of the particular type of distortion existing on the film and to the 
type of analyzer circuit used in making the measurements. 

As was expected, the change from low-gamma negative and white- 
light prints to high-gamma negative with ultraviolet prints has aided 
in increasing the latitude of print density which can be used with an 
acceptable rise in distortion. This increased latitude is a distinct 
advantage, particularly for balancing prints to equal volume and 
for "print up" to obtain high-volume outputs for special sequences. 
It is also important to note that the film distortion as heard appears 
to be less critical to variations of negative density and negative 
gamma than was the case for the normal positive types o! emulsions, 



54 



C. R. DAILY AND I. M. CHAMBERS [J. S. M. p. E. 



this observation being partially explained by the lower minimum 
distortions and the broader tolerance curves 



S/5 



.5 .75 1.0 

LIGHT VALISE Scac 



1.25 1.5 1.75 

, MILS 



FIG. 6. Projected print transmission vs. 
light-valve spacing. 

In conclusion, it can be stated that all the primary objectives of the 
fine-grain program have been accomplished and placed in operation. 




X=V?//vr 



FIG. 7. Intermodulation distortion vs. visual unbiased print 

density. 

As still finer grained films and further improvements in recorder 
optical speed are worked out, greater signal-to-noise ratios may be 



Jan., 1942] FINE-GRAIN FILMS FOR SOUND RECORDING 55 

expected which will in turn allow us to obtain a still more faithful re- 
production of the original sound-source. 

This program has been worked out with close cooperation between 
the film manufacturers, the film laboratory, and the sound depart- 
ment. The writers wish particularly to acknowledge the assistance 
given by Messrs. Wilkinson, Eich, and Gephart of the laboratory 
staff. 

REFERENCES 

1 DAILY, C. R. : "Improvement in Sound and Picture Release through the 
Use of Fine-Grain Film," /. Soc. Mot. Pict. Eng., XXXIV (Jan., 1940), p. 12. 

2 WILKINSON, J. R., AND EICH, F. L.: "Laboratory Modifications and Proce- 
dure in Connection with Fine-Grain Release Printing," J. Soc. Mot. Pict. Eng., 
XXXVHI (Jan., 1942), p. 56. 

8 MILLER, W. C.: "Speed Up Your Lens Systems," /. Soc. Mot. Pict. Eng., 
XXXV (July, 1940), p. 3. 

4 DUPY, O. L., AND HILLIARD, J. K. i "Obtaining Increased Illumination for 
Fine-Grain Film Recording," Amer. Cinemat., 21 (Jan., 1940), p. 36. 



LABORATORY MODIFICATION AND PROCEDURE IN 

CONNECTION WITH FINE-GRAIN RELEASE 

PRINTING * 



JAMES R. WILKINSON AND FERDINAND L. EICH** 

Summary. While fine-grain emulsions have been in general use for specialty 
purposes for three years or more, their use as a medium for release prints is com- 
paratively recent. This paper discusses the necessary modifications required in a 
release laboratory to produce satisfactory fine-grain release prints. The discussion 
covers light-source, power supply, light testing, and printing equipment. Observa- 
tions noted while processing the first thirty million feet of release prints are made 
relative to the behavior and characteristics of the film. 

While a paper on fine-grain film might, at this time, seem somewhat 
dated, in view of the widespread use of this type of emulsion over the 
past three years, if the situation is examined closely it is found that 
fine-grain emulsions, prior to March of the current year, were used 
largely for sound recording and specialty purposes wherein the im- 
proved grain structure has had a beneficial effect upon film noise, 
frequency response, duplicate negatives, master positives, etc. How- 
ever, during the past six months, this great benefit has been extended 
substantially into the release print field. 

The use of Eastman fine-grain positive, type 1302, for release 
prints was of necessity preceded by an extensive program of research 
and testing. These were joint efforts of laboratory and sound engi- 
neers as well as the raw-stock manufacturers, and literally hundreds 
of tests were made using every possible combination of sensitometric, 
processing, exposure, and emulsion variation. From this mass of 
analytical detail were evolved both the final characteristics of the re- 
lease positive material and the optimal specifications for exposing 
and processing the same. 

One of the early determinations in the above program of testing, 
and one which later proved to be a safe anchorage in all calculations, 

* Presented at the 1941 Fall Meeting at New York, N. Y.; received Septem- 
her 30, 1941. 

** Paramount Pictures, Inc., Hollywood, Calif. 

56 



FINE-GRAIN RELEASE PRINTING 57 

was the establishment of a lib gamma of 2.50 =*= 0.10 as an optimal 
area of contrast for the visual observation of prints on the screen. 
All subsequent determinations relative to obtaining optimal sound 
prints, that is, satisfactory frequency response, minimum distortion, 
proper density, gamma, noise level, etc., were subject to reconciliation 
with this predetermined and fixed end point. 

To obtain satisfactory print quality from a variable-density sound 
negative under the above conditions, specifications have been es- 
tablished as indicated in Table I. 

TABLE I 

Negative control gamma 116 . 67 

Negative density . 65 

Print control gamma 116 2 . 50 

Print density 0.58 

Intermodulation distortion 5% 

The above has been covered in detail by C. R. Daily and I. M. 
Chambers. 1 

The first requirement for satisfactory use of fine-grain film was the 
need for sufficient light to print the material. The original experi- 
ment in printing fine-grain emulsions employed the use of mercury- 
vapor high-pressure lamps. These had proved suitable for fine- 
grain master positive printing and had high efficiency in the ultra- 
violet spectrum for sound printing. The levels of light-intensity 
encountered in the use of fine-grain emulsions developed in the Para- 
mount print developer are shown in Fig. 1. This diagram includes 
the present fine-grain release emulsion in use at the present time. 

In order to maintain as flexible a printing system as possible a ceil- 
ing level of light-intensity was adopted which would provide ample 
exposure for master positive emulsions. To secure this, the produc- 
tion printers were equipped with a condenser lens to concentrate the 
light at the back aperture of the Bell & Howell Model D printer. 
To insure a uniform illumination and printing point increment a 
ground-glass was placed between the back aperture and the lens. 
The light-source, which is an AH-8 mercury-vapor lamp, is posi- 
tioned to throw a circular spot of light on the ground-glass. To 
provide easy adjustment to different light-intensity levels for ac- 
commodating the various emulsion types, light attenuators are in- 
serted between the lens and back aperture. These attenuators are 
calibrated so that a level of intensity can be obtained whenever de- 



58 



J. R. WILKINSON AND F. L. EICH [J. s. M. p. E. 



sired without changing the wattage setting of the lamp. Fig. 2 is a 
diagram of the printing set-up. The mercury- vapor lamp is oper- 
ated from a 400-volt, d-c generator, which has a vacuum-tube regu- 
lator for maintaining the supply voltage at =^0.1 volt. The wattage 
of the lamps is adjusted by means of a rheostat in the individual lamp 
circuit to give a standard level of intensity. This brings the wattage 





1 


1 


[ 


]NOTE- WHITE LIGHT INTENSITY 
REQUIRED FOR 
ULTRA-VIOLET PRINTING 


7O 


P|'< 


f 


So, 


jnd 


60 




^ 
















^ 






1 

Ul 










I 








'/ 


z 




















1 




j 






I 










> 30 

5 

cc 




I 






i 






I 










I 






I 






I 






10 




1 






\ 






^ 






^ 










i 






'//, 






//, 






r 







STD. 
POS. 



EXP 
FG. 



MASTER 



RELEASE 
F G 



FIG. 1. 



White light-intensity levels for various printing 
emulsions. 



range of the lamps between 60 and 80 watts. Fig. 3 is the circuit 
used on the production printer lamps. It will be noted that 320 
microfarads of capacitance are placed across the generator output to 
smooth out the voltage ripple. For printing the sound a Corning 
No. 584 filter, 2 mm thick, is inserted between the lens and back 
shutter, and for picture the filter is replaced with a ground-glass. 

In timing the printer exposure, the sound negative density is 
measured on the densitometer and the printer point selected from a 
calibration chart. For picture timing, however, the Cinex tester is 



Jan., 1942] 



FINE-GRAIN RELEASE PRINTING 



59 



retained. This tester is equipped with an AH-8 mercury lamp, 
mounted in a horizontal position in the base of the tester, and pro- 
vides adequate exposure for testing on master positive emulsions. In 
order to drop the intensity level for testing on fine-grain release posi- 
tive, a wire mesh screen is placed between the exposure plane 
and the lamp. The printer-point increment on both the Cinex and 
the printers is maintained at a density of 0.06 per printer point. 

In modifying the Bell & Howell Model 119A automatic release 
printers a more involved change was required to obtain suitable ex- 
posure. The distance between the light-source and exposure plane 
is much greater and the light-beam must pass through a very narrow 



BACK APERTURE 



AH-8 
MERCURY ARC 




LIGHT 
ATTENUATOR 



FIG. 2. Production printer light-source. 



aperture before reaching the exposure plane. Due to the fact that a 
film with a variable-width density is used to control the printer light 
at the narrow aperture, called the mat aperture, the mercury lamp 
is very desirable due to the coolness of the light when concentrated 
at this aperture. The optical system of the printer was altered as 
shown in Fig. 4, to increase the efficiency of the system. This shows 
a side view of the light system. A 25-degree prism is inserted in 
horizontal axis A -A with the base of the prism facing the 90-degree 
prism. Normally with the 90-degree prism in the vertical position, 
the light passing through it would be reflected upward in a vertical 
direction. However, since the exposure aperture is offset from the 
true vertical axis of the 90-degree prism by 25 degrees, as viewed from 
the front of the printer through the plane B-B, the vertical beam of 
light would normally not reach this exposure aperture. In order to 



60 



J. R. WILKINSON AND F. L. EICH [J. S. M. p. E. 



keep all the elements in line on the A- A axis and obtain an offset in 
the beam at the exposure aperture, the 25-degree prism is inserted 
in the optical system and for convenience is placed as indicated in 
Fig. 4. This then rotates the normal vertical axis B-B on the axis 
A- A as a pivot by 25 degrees, and produces a light-image at the ex- 
posure aperture. Since the Model 119- A printers have a lower light 
efficiency than the Model D printers, due to the type of optical sys- 
tem required, the AH-8 lamp was modified to permit higher range of 
wattage. This was obtained by air-cooling the lamp by means of a 



BAL LAST 
RES 




/ 



HAH -s 

LAMP 



FIG. 3. Lamp circuit for production printers. 



blower whose motor is connected across the lamp terminals. This 
cooling system tends to stabilize the lamp and hold it at a constant 
potential. 

An additional piece of optics is necessary in the sound-track side 
as compared to the picture side, since the sound is printed with a 
Corning No. 584 filter, the same as the production sound. In order 
to compensate for the additional loss of light due to the filter, a 
quartz integrating bar is positioned between the mat aperture and 
the exposure plane, to collect all the light and concentrate it at the 
sound-printing aperture. 

By using forced cooling on the lamps the range of wattage under 
which the lamp operates is from 50 to 150 watts. The lamp is main- 



Jan., 1942] 



FINE-GRAIN RELEASE PRINTING 



61 



tained at a voltage of 90 to 100 volts by means of the blower speed, 
and the current in the lamp is adjusted with the lamp rheostat. Fig. 
5 shows the circuit for the release printer lamps. The blower is 
operated with a 220-volt d-c motor connected across the lamp, and 
as the voltage across the lamp varies, the blower speed changes to 
correct the voltage change. With the lamp operating at 100 volts, a 
300-volt d-c. generator is used which provides ample ballast voltage 
for the lamp. An electrical filter is provided in each lamp circuit to 



QUARTZ 

INTEGRATING 

BAR 




TO SUCTION FAN 



FIG. 4. Release printer light system. 



eliminate the voltage ripple from the generator. This is necessary 
because the mercury lamp is stroboscopic in nature and will follow 
slight changes in voltage. 

Fig. 6 shows the lamp assembly for the release printer. The glass 
envelope of the lamp is removed and replaced with a Pyrex chimney, 
which is held in place by the base as shown in the photograph. Fig. 
7 shows the position of the suction fan in relation to the lamp house. 

For starting the lamps a small insulated wire has been located di- 
rectly behind the lamp. A high-voltage spark coil, touched to the 
wire, induces a voltage in the lamp which ionizes the gas and strikes 



62 



J. R. WILKINSON AND F. L. EICH [J. S. M. P. E. 



the arc. This method permits striking the arc without the need of 
opening the lamp house. 

Certain of the major studios divide release printing between West 
Coast and East Coast plants. Due to existing national conditions, 
it has become increasingly difficult for laboratories contemplating 
modification to procure the necessary materials electrical, mechani- 
cal, and optical, to complete their program rapidly. This situation 
has caused some unavoidable delays in obtaining the full benefit of 
the complete fine-grain program. For this interval and until the 
various laboratories have completed their necessary modifications, 
a method has been established, and is in actual practice, whereby 



LAMP BALLAST 




FIG. 5. Lamp circuit for release printers. 



virtually all the improvement embodied in the first three steps of 
sound reproduction with fine-grain film, that is, original recording, 
dubbing prints, and final release sound negative, may be retained 
even though the final release prints are made on conventional posi- 
tives. 

This method comprises the preparation of a photographic duplicate 
sound negative which permits printing by white light to the conven- 
tional positives. From the final release negative a fine-grain master 
positive is prepared. Printing is with ultraviolet light and the mas- 
ter is developed to a gamma of 1.45. From this fine-grain master the 
duplicate negative is printed using another fine-grain material of the 
proper characteristics. Printer exposure is unfiltered mercury light 
and the negative is developed to a density of 0.60 and a gamma of 0.65. 



Jan., 1942] 



FINE-GRAIN RELEASE PRINTING 



63 



This fine-grain duplicate negative may then be printed with normal 
white-light exposure to the conventional positives with an overall 
loss of only 1.5 db at 7000 cycles and with a quality of reproduction 
that is very slightly inferior due to the final release print medium. 
Fig. 8 shows the comparison between distortions of the dupe versus 
the original prints. 

This procedure sacrifices completely the great benefits of fine-grain 
quality to the picture, but it does 
retain the major portion of the 
improvement which has been engi- 
neered by the sound division. 

Prior to the use of fine-grain posi- 
tive for release purposes, there had 
been no practical study, nor was 
knowledge available, relative to 
the effect upon developing solutions 
of large and continuous footages of 
this type of emulsion. It was with 
considerable apprehension and many 
anxious moments that the first few 
days of three to four hundred 
thousand feet per day were care- 
fully checked. However, visions of 
developer imbalance and weird re- 
plenishment requirements due to 
obscure and unanticipated chemi- 
cal reactions proved to be quite 
groundless. Slight changes were 
noted but they were moderate in 
proportion and orderly in their 
progression, and were easily ad- 
justed to permit the maintenance 
of print specifications which had been carefully established. 

Previous development of moderate footages had indicated no 
change would be required in the developing formula. This has like- 
wise proved to be correct with heavy footage and, beyond a slightly 
higher ratio of hydroquinone plus a moderate increase in the H value 
to maintain the correct gamma, the formula has remained unaltered. 
The emulsion does not carry off the same quantity of developing 
solution as did former positives, consequently less actual replenisher 




FIG . 6 . Lamp assembly for release 
printer. 



64 J. R. WILKINSON AND F. L. EICH [J. S. M. p. E. 

is required to maintain the solutions. The emulsion, being somewhat 
thinner than former positive emulsions, dries with considerably less 
heat in the drybox, and due to this characteristic is ejected in a rib- 
bon-like flat condition with no inclination to curl. Some difficulty 
has been encountered in processing the emulsion following develop- 




FIG. 7. Showing position of suction fan in re- 
lation to lamp house. 



ment, and adjustments were necessary in both the processing solu- 
tion and in the method of application to insure a well lubricated sur- 
face. The emulsion has much less pull to its surface than former 
emulsions, and due to its extreme smoothness the surface was, at 
first, non-absorbing and retained an insufficient amount of the process- 
ing fluid. 



Jan., 1942] 



FINE-GRAIN RELEASE PRINTING 



65 



As to uniformity, the stock is equal if not actually superior to the 
former positive. Roll to roll tests on an emulsion of two million feet 
show extremely slight variation and there has been practically no 
drifting tendency either in gamma or density. 




ORIGINAL NEG 
DUP.- 
V^-0.67 Dn-0.63 



!^STER PRINT 
DUP. 228 
\-!36 D,-0.8: 



DUPE NEG. 
DUP. 222 
<-065 D,-0.6' 



PRINT FROM DUPE 
E.K. 1301 



PRINT DENSITY - VISUAL DIFFUSE 



FIG. 8. 



Intel-modulation distortion curves for (a) print from dupe 
negative, and (6) print from original negative. 



Viewing the entire fine-grain program in retrospect, a tremendous 
amount of effort and expense has been necessary to bring it to its 
successful conclusion. As in every worthwhile undertaking prob- 
lems were encountered and their solution was, at times, difficult. 
Nevertheless, these problems have all been well within the bounds 
of reason and the improvement which has been achieved in sound 
and picture quality is ample reward for the efforts expended. 

REFERENCE 

1 DAILY, C. R., AND CHAMBERS, I. M.: "Production and Release Applications 
of Fine-Grain Films for Variable-Density Sound Recording," /. Soc. Mot. Pict. 
Eng., XXXVIII (Jan., 1942), p. 45. 

DISCUSSION 

DR. FRAYNE: The improvement in the appearance of the fine-grain release 
prints appears to have been accompanied by no increase in noise from the sound- 
track. This would seem to indicate that there had been no increase in grain 
size in the more recent fine-grain emulsions offered to the motion picture industry. 
It is my understanding that at least one of the film companies has succeeded in 
removing the brownish color of the image by the addition of a blackening agent 
which results in improved picture quality and does not in any way interfere with 
the sound quality obtained from these types of emulsions. 



A NOTE ON THE PROCESSING OF EASTMAN 1302 FINE 
GRAIN RELEASE POSITIVE IN HOLLYWOOD* 

V. C. SHANER** 



Summary. A brief historical resume is given of a series of fine-grain films that 
have been put upon the market during the past four years. This series of fine- 
grain films culminated with the acceptance of Eastman 1302 fine-grain release posi- 
tive at one Hollywood laboratory to the exclusion of regular positive of the 1301 type 
for release printing. Experimental data are presented to show the comparative 
sensitometric characteristics of fine-grain positive 1302 and regular positive 1301 
at various pH values and potassium bromide concentrations typical of Hollywood 
positive developers. A basic positive developer formula derived from chemical 
analyses of every release positive developer in Hollywood was used in the experi- 
mental work. Some practical facts are discussed, based upon the experiences ob- 
tained from the initial use of the fine-grain film in Hollywood. 

The introduction during the year 1941 of Eastman Fine Grain 
Release Positive, Type 1302, marked the completion of a series of 
fine-grain films which have been made available to the motion pic- 
ture industry during the past several years. Before discussing this 
new film and its processing in detail, it may be well to review briefly 
the preceding fine-grain products. 

In 1937, two fine-grain duplicating materials 1 were introduced to 
meet the need of eliminating excessive graininess in prints made 
from duplicate negatives. These two films, Eastman Fine Grain Du- 
plicating Positive, Type 1365, and Eastman Fine Grain Panchromatic 
Duplicating Negative, Type 1203, won ready acceptance and today 
are standard products in the duplicating field. The year 1938 
marked the appearance of Eastman Fine Grain Sound Recording 
Film, Type 1360, the first of several emulsion types to bear this name. 
Designed for use in variable-area recording, it gave improved high- 
frequency response together with a definite decrease in ground-noise. 
The picture-negative films had their major improvement from the 
standpoint of fine-grain characteristics in 1938 also, and in that year 

* Presented at the 1941 Fall Meeting at New York, N. Y.; received October 
10, 1941. 

** Eastman Kodak Co., Hollywood, Calif. 
66 



PROCESSING OF EASTMAN 1302 POSITIVE 67 

Eastman Plus-J^f, Type 1231, was first manufactured and submitted 
to the trade. 2 This film has twice the speed of Eastman Super- X, 
Type 1227, its predecessor as standard production negative film, yet 
is definitely finer in grain. In 1939, another Fine Grain Sound Re- 
cording Film, Type 1366, brought to the field of variable-density re- 
cording the improved sound quality and reduced ground-noise which 
the 1360 type had made possible in variable-area recording. It is 
interesting to note here that research on fine-grain sound-recording 
materials has not ceased, and even today new experimental films are 
continually being developed and tried out in production. 

The present year, 1941, has witnessed the introduction of Eastman 
Fine Grain Release Positive, Type 1302, and its adoption to the ex- 
clusion of the regular Positive, Type 1301, in the production work of 
one major Hollywood release laboratory, namely, Paramount. 
Other laboratories in this area have also made some production use of 
the 1302 type. It is in this new fine-grain release print material 
that the complete effects of the other fine-grain materials are realized 
and extended to the theater audience. 

In the period of months since the use of the Fine Grain Release 
Positive began in Hollywood, approximately 50,000,000 feet have 
been consumed in release print manufacture. The result, a finer 
grained image on the theater screen, has been attained not only 
through the contribution of the film manufacturer in developing a 
better product, but also through the ability of the motion picture 
laboratory technicians to work out the problems involved in the use 
of the new film. 

The manufacture of a film with finer-grain characteristics neces- 
sitated a sacrifice of emulsion speed, with the result that the Fine 
Grain Release Positive has approximately one-quarter the emulsion 
speed of the regular 1301 Positive. Laboratory technicians were, 
therefore, confronted with the major problem of readjusting the ex- 
posure level of production printers. Some laboratories solved this 
problem satisfactorily by equipping the printers with mercury-vapor 
lamps, which emit more radiation than tungsten lamps in the shorter 
wavelengths, to which the film is most sensitive. Other laboratories 
installed high-intensity tungsten lamps and, in some instances, used 
reflectors to increase the effective exposure. Thus the Hollywood 
laboratories have modified their printing equipment in one way or 
another to accommodate the lower-speed film. 

There is another characteristic of the Fine Grain Release Positive 



68 



V. C. SHANER 



[J. S. M. p. E. 



which requires consideration, although it did not necessitate so great 
an alteration in production laboratory technic as did the lower emul- 
sion speed. If two prints from the same negative, one on regular 
Positive, Type 1301, and the other on Fine Grain Release Positive, 
Type 1302, are matched for density and contrast when viewed by 
diffused light over an illuminator, then these same prints, when pro- 
jected, will no longer match in quality. The 1302 print will be 
lighter in density than the 1301 print, and will have lower contrast. 
These effects are due to differences in the optical characteristics of 
the silver particles in the two films. 





FIG. 1. Distribution of light flux through the two film types used in projection. 



In Fig. 1, polar diagrams illustrate the distribution of light flux in 
the beam of light transmitted through the two film types under con- 
ditions similar to those encountered in projection. It is readily ap- 
parent that the smaller silver particles of the fine-grain print produce 
less scattering of the transmitted light. Thus, although the prints 
appear equal in density and contrast when viewed by diffused light 
over an illuminator, under the conditions of specular illumination 
which prevail in a projector, the lens receives more light from the 
fine-grain print. The result is lighter and less contrasty screen 
quality. 

Experience has shown that in addition to the increase in exposure 
required by the lower speed of the 1302 emulsion, a further increase 
of approximately two Bell & Howell printer steps is required to bring 



Jan., 1942] PROCESSING OF EASTMAN 1302 POSITIVE 



69 



the screen density of a 1302 print up to the screen density of a 1301 
print. To compensate for the lower contrast seen on the screen, the 
1302 print must also be developed to a higher gamma, approximately 
2.50, as contrasted with the normal 1301 gamma of 2.10. With 
these changes, projected prints on the two materials can be matched 
for density and contrast. This does not mean, however, that they 
are equal from the standpoint of general photographic quality. Ac- 
tually, the 1302 print is of much higher quality because the lower 
graininess produces a smoother, rounder appearance of the photo- 
graphic image on the screen. 



3.20. 

3.00. 

2.80. 

2.6a 

2 2.40. 

<2.2d 

2.00. 

1.80. 

1.60 






2.10 



2468 02468 024 

TIME OF DEVELOPMENT IN MINUTES 



6 8 



FIG. 2. Time of development vs. gamma; X-2 developer; 
(1} 1302; (2) 1301. 



Generally speaking, fine grained materials have rapid rate-of-de- 
velopment characteristics; that is, the contrast increases rapidly 
with time of development. Thus the development rate of 1302 is 
somewhat higher than that of 1301. In view of the higher contrast 
requirements of 1302 prints, this is a desirable feature, since in many 
cases longer times of development would result in decreased produc- 
tion. 

The time-gamma curves in Fig. 2 show the comparative rates of 
development of 1301 and 1302 in a positive developer adjusted to 
three different pH levels. The developer, designated X-2, is derived 
from chemical analyses of all the positive developer solutions in use 
in Hollywood, using methods previously described. 3 It is employed 
by the Eastman West Coast Laboratory for many test purposes. 



70 V. C. SHANER [J. S. M. P. E. 

The actual constituents, as well as the amounts of each used to make 
up the basic formula, are contained in the following tabulation : 

Elon 1.5 grams 

Hydroquinone 3.0 grams 

Sodium Sulfite, desiccated 40.0 grams 

Potassium Bromide 2.0 grams 

Sodium Carbonate, desiccated 17.0 grams 
Water to make 1.0 liter 

R 10.0 



The curves in Fig. 2 were obtained from data derived from the de- 
velopment of regular Type lib sensitometric exposures. Prior to ex- 
posure, the film samples were conditioned for two hours in an atmos- 
phere of 68 per cent relative humidity at a dry-bulb temperature of 
70 F. The strips were developed at 67 F in a sensitometric develop- 
ing machine previously described in the JOURNAL. 4 

The H values were measured with a Beckman laboratory model 
H-meter equipped with an improved glass electrode which has no 
sodium ion correction in the H range of these tests. Developer 
modifications having pTL values of 9.8 and 10.2 were obtained by 
altering the amount of desiccated sodium carbonate in the X-2 
formula to 7 and 22 grams per liter, respectively. 

At a H value of 9.8, a development time of 4.1 minutes was re- 
quired to obtain a gamma of 2.50 with 1302. At the same time of de- 
velopment a gamma of 2.10 was obtained with 1301. Similarly, at 
pH values of 10.0 and 10.2, development times of approximately 3.0 
and 2.0 minutes gave the desired gamma values with both films. 
Since the H values of 9.8 to 10.2 represent the current working 
range of the positive developers used in Hollywood, it becomes clear 
that 1302 can be developed at machine speeds approximately the 
same as those used with 1301, even though a higher contrast is re- 
quired. 

The chemical analyses of the positive developers used in Hollywood 
laboratories disclosed potassium bromide concentrations ranging 
from 0.75 to 4.0 grams per liter. Although the effects of various con- 
centrations of this chemical on positive film of the 1301 type are 
well known to laboratory chemists, it may be of interest to summarize 
them here. An increase in potassium bromide concentration pro- 
duces a corresponding decrease in density. If the bromide concen- 
tration is increased to approximately 2 grams per liter, gamma in- 



Jan., 1942] PROCESSING OF EASTMAN 1302 POSITIVE 



71 



creases, but if the bromide concentration is increased to more than 2 
grams per liter, the retarding effect on development is sufficient to 
decrease the gamma as well as the density obtained at a given time 
of development. 

Fig. 3 shows the effects of various concentrations of potassium 
bromide on the gamma and density of 1301 and 1302 developed in 
the X-2 developer. It will be noted that bromide has similar effects 



P Hi 10.2 
P Hr9.8 




1302 



^1301 



1301 STEP 1 5 AT 1-. 2.10 

1302 STEP II AT Y* 2.50 



0.5 1.0 2.0 4.0 

KBr IN GRAMS PER- LITER 
FIG. 3. Effect of bromide on gamma and density. 

on the contrast of both films at a pU of 9.8. At a pU of 10.2, how- 
ever, the gamma of 1302 increases with increasing bromide concen- 
tration up to 4.0 grams per liter. Thus, at higher H levels, the de- 
crease in gamma experienced with 1301 at high bromide concentra- 
tions does not appear with 1302. 

Referring now to the density curves, it will be seen that at a bromide 
concentration of 1.0 gram per liter, the densities obtained with 1302 
were 0.93 at a H of 9.8, and 0.80 at a pH of 10.2, while at a concen- 
tration of 4.0 grams per liter the densities were 0.54 at a pH of 9.8 
and 0.43 at a pH of 10.2. To the practical laboratory chemist, these 



72 V. C. SHANER [J. S. M. P. E. 

decreases in density are equivalent to an exposure loss of about five 
Bell & Howell printer steps at a gamma of 2.50. Under correspond- 
ing conditions, the densities obtained with 1301 were 0.78, 0.70, 0.58, 
and 0.50, respectively. Here the decreases in density are equivalent 
to an exposure loss of approximately three Bell & Howell printer 
steps at a gamma of 2.10. Since the exposure loss is two printer 
steps less than that experienced with 1302, it is apparent that the 
rate of decline of density with increasing bromide concentration is 
somewhat greater for the fine-grain film. 

Objectionable fog values were encountered on both emulsion types 
in X-2 developer containing no potassium bromide. However, the 
fog values obtained in the absence of bromide were lower on 1302 
than on 1301. 

These data have been presented to illustrate the characteristics 
of Eastman Fine Grain Release Positive Film, Type 1302, under the 
conditions existing in Hollywood release print laboratories. The 
use of radically different developer formulas may lead to conclusions 
at variance with those expressed here, and in some cases it may be 
necessary to modify such formulas in order to secure the best results. 
It is felt, however, that the information given here should provide a 
basis for activities with the new film outside the Hollywood area. 

REFERENCES 

1 CRABTREE, J. I., AND IVES, C. E.: "Two New Films for Duplicating Work," 
J. Soc. Mot. Pict. Eng., XXIX (Sept., 1937), p. 317. 

2 HUSE, E., AND CHAMBERS, G. A.: "Three New Eastman Negative Emul- 
sions," Amer. Cinemat. (Dec., 1938), p. 187. 

3 ATKINSON, R. B., AND SHANER, V. C.: "Chemical Analysis of Photographic 
Developers and Fixing Baths," /. Soc. Mot. Pict. Eng., XXXIV (May, 1940), 
p. 485. 

4 JONES, L. A., RUSSELL, M. E., AND BEACHAM, H. R.: "Machine for Sensi- 
tometric Work," J. Soc. Mot. Pict. Eng., XXVIII (Jan., 1937), p. 73. 

DISCUSSION 

MR. GASKI: How did you arrive at the X-2 formula for 1302? 

MR. HUSE : For some period of time a number of samples of actual machine 
positive developers were taken from each of the various laboratories in Holly- 
wood. Chemical analyses of these various developer samples were made. Fol- 
lowing this, an average formula was derived from these analytical data, which 
formula in a sense represented the numerical average of all. 

Numerous photographic tests later proved that this formula gave us a good 
working developer for correctly analyzing the behavior of films in terms of 
laboratory practice. 



Jan., 1942] FlNE-GRAIN RELEASE PRINTING 73 

MR. HYNDMAN: The majority of laboratories in New York City and the 
surrounding area have made printing and developing tests on fine-grain release 
positive films. Consolidated and DeLuxe Laboratories have printed complete 
releases on this material. Within thirty days other laboratories will be ready to 
do release printing on this film and probably within the next thirty to ninety 
days several laboratories will do all their release printing on this film. 

Most of the eastern laboratories have found it necessary to revamp their 
printers so as to obtain sufficient exposure to print this film. A number of 
laboratories have also made changes in their positive developer bath to secure 
what was considered the best photographic quality. 

MR. GRIFFITH: Why, in the processing of fine-grain film, is less developer 
carried out of the developer tank by the film itself? 

MR. HUSE: The physical characteristics of 1302 are such that less developer 
is actually absorbed per unit length or unit area of the film during processing. 



REPORT OF THE THEATER ENGINEERING COMMITTEE 



Summary. This is an account of the work of the several sub-committees of the 
Theater Engineering Committee during the past two years. The report of the Sub- 
Committee on Projection Practice embodies a preliminary study of safety factors in 
projection rooms, specifically with reference to the use of hand-operated fire extinguish- 
ers. 

The report of the Sub-Committee on Theater Design includes a preliminary study 
of the basic shapes of theaters and advantageous seating zones, including a report on a 
study of these factors made in the Surrey Theater in New York. 

A report of the Sub- Committee on Screen Brightness deals with proposed specifica- 
tions for meters for measuring incident and reflected screen light in theaters, and the 
efforts of the Committee to encourage the manufacture of such instruments for the 
industry. 

In a previous report of the Committee, published in the December, 
1940, issue of the JOURNAL, announcement of the reorganization of 
the Theater Engineering Committee was first made. Originally, 
the Committee consisted of the two Sub-Committees on Projection 
Practice and Theater Design. Under the new arrangement, a third 
sub-committee took over the work dealing with screen brightness, 
leaving the Projection Practice Sub-Committee to concentrate its 
attention on projection problems proper. 

The personnel of the Theater Engineering Committee, divided 
into the three sub-committees, is as follows : 

THEATER ENGINEERING COMMITTEE 

ALFRED N. GOLDSMITH, Chairman 



Projection Practice Sub-Committee 




H. RUBIN, Chairman 




H. ANDERSON 


E. R. GEJB 


P. J. LARSEN 


T. C. BARROWS 


M. GESSIN 


J. H. LlTTENBERG 


H. D. BEHR 


A. GOODMAN 


E. R. MORIN 


K. BRENKERT 


H. GRIFFIN 


J. R. PRATER 


F. E. CAHILL, JR. 


S. HARRIS 


F. H. RICHARDSON 


C. C. DASH 


J. J. HOPKINS 


J. J. SEFING 



* Presented at the 1941 Fall Meeting at New York, N. Y.; received October, 
15, 1941. 

74 



THEATER ENGINEERING COMMITTEE REPORT 



75 



A. S. DICKINSON 
J. K. ELDERKIN 
J. FRANK, JR. 
R. R. FRENCH 



F. W. ALEXA 
D. EBERSON 
J. FRANK, JR. 
M. M. HARE 
S. HARRIS 



F. T. BOWDITCH 
W. F. LITTLE 



C. HORSTMAN 

L. B. ISAAC 
I. JACOBSEN 

Theater Design Sub-Committee 

B. SCHLANGER, Chairman 
C. HORSTMAN 
E. R. MORIN 

K. C. MORRICAL 

I. L. NIXON 

Screen Brightness Sub-Committee 

F. E. CARLSON, Chairman 
S. HARRIS 
W. B. RAYTON 
C. TUTTLE 



R. O. WALKER 
V. A. WELMAN 
H. E. WHITE 
A. T. WILLIAMS 



C. C. POTWIN 
A. L. RAVEN 
R. F. Ross 
E. S. SEELEY 
J. J. SEFING 



H. E. WHITE 
A. T. WILLIAMS 



Sub-Committee on Projection Practice 

Working Committee on Fire Hazards. A special subject before the 
Working Committee on Fire Hazards dealt with the question of in- 
cluding hand-operated fire extinguishers as part of the equipment 
of the motion picture projection room. The first step in the study 
was to send communications to various manufacturers of hand-oper- 
ated fire extinguishers, stating the problem facing the Committee 
and asking specifically the following questions : 

(1) What effect has your extinguisher on burning film, especially of the cellu- 
lose nitrate type? 

(2} What damage to other equipment in the projection room might be 
incurred from the use of the extinguisher? 

(5) What, if any, possibly toxic vapors are produced from the use of the ex- 
tinguisher on burning cellulose nitrate film, and in what amounts? 

In view of the inadequacy of information pertaining to these sub- 
jects, no definite answers to these questions were available either from 
the manufacturers of the equipment or from the information at the 
command of the Committee. However, the questions indicate 
some of the important data that should be obtained by the Com- 
mittee aside from the question of establishing a policy with regard 
to the use of hand-operated extinguishers in projection rooms. 

A reply received from Mr. R. M. O'Connell, Service Engineer of 
Underwriters Laboratories, Inc., Chicago, contains paragraphs of 
special interest to the Committee, since the thoughts expressed agree 



76 THEATER ENGINEERING COMMITTEE REPORT [j. s. M. P. E. 

strongly with the feelings of the Projection Practice Committee ex- 
pressed frequently in previous reports and at many meetings of the 
Committee. These paragraphs follow : 

We also attach a copy of the Regulations of the NBFUfor Nitrocellulose Motion 
Picture Film. Section 19 of this pamphlet is intended to afford necessary safe- 
guards for booths, including vents, shutters, and noncombustible construction. 
We may call your attention to the note following sub-paragraph J appearing on 
p. 22 of this pamphlet. This note recommends the installation of automatic 
sprinklers wherever practicable. 

In our study of the subject we have come to believe that the fundamental pur- 
pose of the above Regulations is to afford protection to the other parts of the 
building and to the occupants rather than to suggest means of controlling any 
film fires which may actually occur within the projection room. As you know, 
such fires burn rapidly, give off intense heat, and great volumes of suffocating 
fumes and in our opinion ordinarily could not be controlled by hand-operated 
extinguishers of the usual type, even though it were possible for the occupants 
of the projection room to put such extinguishers into action and remain within 
the booth for any appreciable time following the start of a fire. As you know, 
nitrocellulose film is not dependent upon supplies of oxygen from the surrounding 
atmosphere. The entire intent of the Regulations therefore seems to be that the 
operator should try to get out of the room or booth as quickly as possible and 
hope that the booth itself was so constructed and ventilated that the film fire 
would burn out without extending into the building and without emitting a haz- 
ardous volume of fumes to the rest of the surroundings. 

In our opinion the safety of booths can not be made dependent upon hand fire 
extinguishers. It would, of course, be well to have proper extinguishers close at 
hand outside of the booth in case a need for them should arise. You will observe 
in Section 14, Rule 144, of the National Board Regulations, a note which recom- 
mends small hose equipment and extinguishers except in film vaults. This is 
probably a reasonable recommendation, but we would not depend too much on 
extinguishers to handle film fires unless of the very smallest size and only if the 
extinguishers were brought into action quickly before very much film was in- 
volved. It is probably more true of films than of other combustibles that protec- 
tion is to be sought in preventing fires rather than by provision for extinguishing 
them after they have once started. 

The Chairman has obtained permission of Mr. O'Connell to quote 
these paragraphs in a report of the Committee. 

Another question considered by the Committee was a possible 
inconsistency between Sections 144 and 218 of the Regulations of the 
National Board of Fire Underwriters for Nitrocellulose Motion Picture 
Film as recommended by the National Fire Protection Association 
and published by the National Board of Fire Underwriters as NBFU 
pamphlet No. 40. 



Jan., 1942] THEATER ENGINEERING COMMITTEE REPORT 77 

Section 144 reads as follows : 

Every room in which film is stored or handled, except film vaults, shall be 
provided with first aid fire appliances of types using water or water solutions. 
(Then follows a list of several extinguishers considered suitable.) 

Section 218 reads as follows: 

In the event of film fire in a projector or elsewhere in a projection or rewind 
room, the projectionist should immediately shut down the projection machine 
and arc lamps, operate the shutter release at the nearest point to him, turn on the 
auditorium lights, leave the projection room, and notify the manager of the theater 
or building. 

It was pointed out that if the projectionist should leave the pro- 
jection room in the event of a fire, there would be no point in having 
hand-operated extinguishers inside the projection room. 

The Committee felt that a hand extinguisher might perhaps be of 
use in cases of small fires from sources other than film, but in turn 
it was pointed out that nothing that would be likely to burn was per- 
mitted in the projection rooms, according to the Regulations. 

There is apparently no definite information concerning cases of 
fire where hand extinguishers have been used, and most of the in- 
formation available with regard to film fires in the projection room 
and the extinction of such fires is incomplete and sometimes ques- 
tionable. 

The general consensus of the Committee may be summed up as 
follows: 

(1) The Committee felt that no hand-operated extinguishers should be in the 
projection room. 

(2} One or more hand extinguishers should be available immediately outside 
the door or doors of the projection room. 

(5) The Committee still feels that in the event of film fire, the projectionist 
should immediately leave the projection room, so that Section 218 of the Regula- 
tions is to be regarded as satisfactory, but that Section 144 should either be 
omitted entirely or revised in accordance with Items 1 and 2. 

Working Committee on Tools and Tolerances. It is the convic- 
tion of the Projection Practice Sub-Committee that progress in the 
projection art requires that there be promptly made available, prefer- 
ably by the projector manufacturers, information on exact methods 
and appropriate tools for measuring the wear of projector parts, data 
on the permissible maximum tolerable amount of wear of each part 
before required replacement, accurate methods of measuring such op- 



t8 THEATER ENGINEERING COMMITTEE REPORT [J. S. M. P. . 

erating values as film tension at the gate, and the corresponding 
convenient tools for measuring and adjusting such operating condi- 
tions. 

The Sub-Committee points out that it has endeavored for a period 
of years to secure such information and tools, and at this time regards 
the lack of such material as detrimental to the advancement of pro- 
jection and accordingly urges the early availability of such data and 
tools. 

Sub-Committee on Theater Design 

The Theater Design Sub-Committee is endeavoring to formulate 
plans for a series of surveys from which information may be derived 
which will indicate those zones of seating in the motion picture audi- 
torium most preferred for comfortable viewing of the pictures. It is 
intended also to locate the zones of second, third, and even lesser 
choice for seating as selected by the audience after the more highly 
preferred areas are filled. 

The Committee fully realized that there would be many significant 
factors which might change the pattern of the preferred zones; for 
example, the size and brightness of the picture might have a direct 
influence on the pattern. Also, the traffic lines into the auditorium 
and the placing of the aisle leading to the seats would be relevant 
factors. While it is realized that poor sightlines due to improperly 
pitched floors and uncomfortable chairs might influence the location 
of preferred seating zones, the Committee feels that it would be wise 
to place little stress on these last two factors because theaters having 
such conditions could and probably should be avoided for this survey 
work. 

To arrive at any worthwhile conclusions, it is felt that it would be 
necessary to make surveys for auditoriums of varied basic shapes, such 
as the square shape, the extremely elongated rectangle, and in-be- 
tween shapes. It would also be necessary to survey theaters of varied 
capacities, the 600, 900, 1200-seat capacities being recommended 
for the tests. It would be preferable if the size and brightness of the 
picture could be varied for a given seating pattern so that their in- 
fluence could be more definitely observed. 

The first survey was made in a theater under actual operating 
conditions. This type of survey can be made fairly accurate and 
gives a true picture of the preferred seating zones. The main dif- 
ficulty, however, arises from the fact that it is necessary to have at 



Jan., 1942] THEATER ENGINEERING COMMITTEE REPORT 79 

least one checker for approximately every 60 chairs in an auditorium. 
Each checker must have a chart in front of him indicating the chairs 
in his zone so that he can mark the chairs as they become occupied. 

Still another method of checking the preferred seating zones was 
considered in which the audience would be brought to an auditorium 
chosen for test purposes, and, under different conditions be asked to 
seat themselves in accordance with their ideas of comfortable viewing 
positions. It is probably true that an unsuspecting audience would 
give more conclusive information but there are definite advantages in 
this latter type of survey. It would not be necessary to have large 
squads of checkers. It must be remembered that considerable travel 
and arranging of available time is involved in getting together a group 
of checkers to make a survey in theaters in actual operation. In the 
plan which involves the use of an audience chosen for the test purpose, 
the necessary changes in picture size and brightness could be made and 
their effect noted on the specific audience. It would also be possible 
to rope off designated seating areas in which the basic shape of the 
seating pattern could be varied to check on the influence of the basic 
shape. Of course, one major obstacle to this type of survey would be 
the difficulty of obtaining an auditorium equipped with the neces- 
sary chairs and projection equipment for the tests. The difficulty of 
obtaining a sufficiently large audience could be minimized by spacing 
the chairs farther apart in both directions than would be normal prac- 
tice so to decrease the required number of viewers ; for example, the 
normal audience of 600 could be tested with approximately 200 persons. 

The Committee made an actual survey in the Surrey Theater in 
New York City. The theater was open for business and the tests 
were made starting at 6:45 P. M. and ending at 9 P. M. The theater 
has a capacity of 570 chairs, and in the hours indicated 453 people 
entered the theater to view the screen performances. The accom- 
panying diagram indicates the plan of the theater, the position of the 
screen, and the like. The results of this survey are herewith given 
and the sub-committee may continue with a series of these surveys un- 
less it is found that other more practicable methods can be used 
to arrive at the necessary results. 

The survey shown in Fig. 1 was taken, as stated, at the, Surrey 
Theater located in the Bronx, New York. Eleven members of the 
Sub-Committee entered the theater on a week-day evening at about 
6:30 P. M. Each sub-committee man occupied a specified seat from 
which he could view an area of approximately 50 seats. The survey 



80 



THEATER ENGINEERING COMMITTEE REPORT [j. s. M. p. E. 



started at 6:45 P. M. and terminated at 9 P. M., the evening period in 
which the major part of the audience was expected to arrive. The 
total period was divided into three periods: 6:45 P. M. to 7:30 P. M. 
for the first period; 7:30 P. M. to 8:15 P. M. for the second period; 
and 8:15 P. M. to 9 P. M. for the third period. Each man had a seat- 
ing chart in front of him which enabled him to record those seats 
which were occupied in the first, second, and third periods. 

Fig. 1 shows the seating diagram of the theater and black boxes of 
various sizes indicate in which period the chairs were occupied. Fig. 1 
shows the different weights given to the black markings in accordance 



l~] AREA FOR ONE SEAT 

HD OCCUPIED IN FIRST PERIOD 
C3 - M SECOND 

1~3 > .. THIRD 



5- 
o < 




54321 

FIG. 1. Theater seating chart for determining preferred seating areas. 



with the period when the occupancy occurred. Greater weight was 
given to the earlier periods so that a visual picture could be obtained 
of the preferred seating locations. It also assumed, of course, that the 
chair locations occupied in the earliest period would indicate the 
highest preference. 

In this particular theater the facts that the approach of traffic 
was from, one side rather than from the usual center approach, and 
that the smoking section was placed to one side threw the weight of 
preferred seats to one side, as the chart indicates. The picture size 
in this theater was 12 ft 7 in. X 17 ft 5 in. and the screen illumina- 
tion was a little above the average in intensity. The picture size was 
larger than the average size to be expected for the maximum viewing 



Jan., 1942] THEATER ENGINEERING COMMITTEE REPORT 81 

distance of this theater. The maximum viewing distance was 4.85 X 
the picture width. In accordance with a previous survey made by 
this Committee, the average picture size in relation to the maximum 
viewing distance was found to be the maximum viewing distance 
divided by 5.2. It is not assumed at this time that this single survey 
could by any means give conclusive information as to preferred seat- 
ing arrangements. It would be necessary to make approximately a 
dozen or more of these surveys under different conditions as already 
suggested in this report. However, it is interesting to note some of 
the disclosures made by this survey. These are as follows : 

(1) That seating locations in an area near the picture starting 
with the picture and ending with a distance approximately I 1 / \ 
times the picture width away from the picture, are resorted to only 
very infrequently. 

(2) That the preferred viewing distances from the picture are 
found in an area located at distances beginning at approximately 
twice the picture width and ending at approximately four times the 
picture width. Fig. 1 is marked with a scale at the bottom to show 
the relation of viewing distances to the picture width. Each unit 
marked on the scale is equal to the picture width. 

(5) That seats located in an area too far to one side of the pic- 
ture, or such as may be located outside an angle of approximately 
60 in relation to the picture surface as shown on the chart, are not oc- 
cupied any sooner than the two seats heretofore mentioned in the 
front sections, when other seats are available. 

Conditions in this particular theater did not permit any worthwhile 
observations to be made as to what could be considered as useful 
seating areas at more remote distances from the picture. 

Sub-Committee on Screen Brightness 

As reported at the last Convention of the Society, the Sub-Com- 
mittee formulated provisional specifications for illumination meters 
and brightness meters and was placing these before instrument manu- 
facturers to determine the feasibility of having them made available. 
It is proposed to review the results of this investigation in this report. 
Before doing so, however, it seems desirable to summarize briefly the 
preliminary conclusions and opinions of the Sub-Committee on which 
the provisional specifications were based. 

The Sub-Committee has felt that instruments for merely measuring 
total flux in the beam from the projector, or average brightness of 



82 THEATER ENGINEERING COMMITTEE REPORT [j. S. M. p. E. 

the screen, would be inadequate and that the instruments should be 
of such a character as to permit determinations of uniformity of il- 
lumination and brightness. This introduces no serious problems so 
far as illumination meters are concerned, but it does mean that, in 
the case of the proposed brightness meter, the instrument must be de- 
signed so that it "sees" only a limited area of the screen at a time. It 
was also agreed that, if the brightness determinations are to represent 
a true measure of what the audience sees, the measurements should 
be made from the seating area and include the extreme seat positions. 




FIG. 2. Suggestion for possible form and use of 
illumination meter. 

Data presented by the Projection Practice Committee in 1938 
indicate a minimum angle subtended by the screen of approximately 
8 degrees. Therefore, the "angle of view" of the brightness meter 
should preferably be not more than 2 degrees if a reasonable indica- 
tion of brightness uniformity is to be obtained from the most distant 
seats. 

If possible, the use of "visual" types of instruments should be 
avoided because of the difficulty of matching brightness fields with 
color differences present. The use of photoelectric cells of either the 
photoemissive or photovoltaic types corrected to eye sensitivity was 
recommended. 



Jan., 1942] THEATER ENGINEERING COMMITTEE REPORT 83 

Since measurements of illumination at a variety of points on the 
screen area are required to obtain a reasonably accurate average 
value, the illumination meter should be separate from the photocell 
and connected to it by a suitable length of conductor. Similarly, 
means should be provided to hold the cell at any point on the screen 
area without resorting to the use of ladders or other cumbersome 
equipment. 

Since measurements of brightness at a variety of areas on the 
screen were proposed, at least the "viewing" part of the brightness 
meter should be swivel mounted on a suitable support, or tripod, and 
means provided for aiming at specific points on the screen. Also, 
the aiming device, or "view finder," should probably indicate spe- 
cifically the field included by the instrument. 

To clarify the several points summarized here and included in the 
letter circulated to instrument manufacturers, two sketches were 
prepared to illustrate the proposed use of the meters (Figs. 2 and 3). 
The letter also included a tabulation of the provisional specifications 
(Table I). 

TABLE I 

Provisional Specifications for Illumination and Brightness Meters 

Illumination Brightness 

Meter Meter 

Useful Range of Instruments 0.2-50 ft-candles 0.5-30 ft-lamberts 

(or 4-30) 

Accuracy of Measured Values =fc 5% =b 5% 

Reproducibility of Measured Values =t 3% =^3% 

Max. Screen Area or Angle to be In- 
cluded by Instruments 1 sq ft 3 (preferably 2) 

Maximum Price $50 $50 



The Sub-Committee's choice of readable brightness values from 
0.5 to 30 ft-lamberts was included provisionally to permit measure- 
ment of the brightness of the peripheral field should that area be il- 
luminated. The alternative range of values from 4 to 30 ft-lamberts 
was included if it should be found impracticable to obtain an in- 
strument capable of reading 0.5 ft-lambert. 

The values for the illumination meter of from 0.2 to 50 ft-candles 
were chosen in recognition of the requirement in some states of 
minimum ambient levels of illumination in theater auditoriums, and 



84 



THEATER ENGINEERING COMMITTEE REPORT [j. S. M. p. E. 



because of the numerous other illumination measurements for which 
such an instrument could be used in theaters. 

The letter to instrument manufacturers also included a request 
to consider the practicability of an inexpensive amplifier for increas- 
ing the sensitivity of brightness meters employing photocells. Since 
the frequency of the light on a motion picture screen is 48 cycles per 
second, it was felt that perhaps a 48-cycle amplifier, with a null 
method of indication, might have advantages. 

The general letter outlining the problem and the provisional speci- 
fications was sent to 30 instrument manufacturers. Replies were 



/tewing part ol brightness meter 

2 to 3" held of viev 




FIG. 3. Suggestion for possible form and use of bright- 
ness meter. 

received from 20 of these companies. Twelve of them expressed either 
no interest in the subject or an inability to take action at the present 
time. The remainder expressed a willingness to cooperate and several 
submitted helpful suggestions and criticisms. As a result, the Sub- 
Committee first of all revised the tentative specifications for accuracy 
and reproducibility of measured results from 5 to 15 per cent and from 
3 to 5 per cent. 

As anticipated, the attainment of a satisfactory illumination meter 
seems to present no serious problems. However, the design of the 
desired type of brightness appears considerably more difficult and the 
Sub-Committee feels that it should devote its attention to that prob- 
lem until a practicable design has been found. 



Jan., 1942] THEATER ENGINEERING COMMITTEE REPORT 



85 



I 



Subsequent calculations by members of the Sub-Committee indi- 
cate the magnitude of the brightness meter problem. Assuming the 
use of a collector lens which would also serve to limit the field of view 
of an instrument employing a photoelectric cell, the illumination at 
the lens with 4 ft-lamberts on the screen in the 2-degree angle would 
be 1 X 10~ 3 ft-candle at any distance from the screen. 

A photovoltaic cell corrected to eye sensitivity and used as pro- 
posed would have an output of the order of one or two thousandths 
of a microampere under these 
conditions. Such currents are 
too low to read on any con- 
venient, rugged form of instru- 
ment and, therefore, the Sub- 
Committee is extremely doubtful 
of the practicability of using such 
cells, unless the tentative specifi- 
cations are radically altered at 
the expense of many features 
now considered to be eminently 
desirable. 

The use of the photoemissive 
type of photoelectric cell presents 
many difficulties. Under the 
conditions previously described, 
an output from the cell of the 
order of 2 X 10~ 5 microampere 
seems likely. While so low an 
output does not necessarily rep- 
resent an insurmountable ob- 
stacle, it should be evident from 
the foregoing data that the design of a suitable brightness meter em- 
ploying either type of photocell would require a considerable amount 
of development work. 

An interchange of ideas between the Sub-Committee and those in- 
strument manufacturers interested in a photocell type of brightness 
meter continues. In the meantime, however, the Sub-Committee is 
reconsidering visual types of brightness meters, since one such in- 
strument is already available (Fig. 4) at a price close to the figure con- 
sidered to be acceptable. 

One objection to a visual type of instrument has been the difficulty 




FIG. 4. A simple brightness meter 
employing a light-meter for the in- 
dicating instrument. 



86 THEATER ENGINEERING COMMITTEE REPORT 

of matching fields of brightness when a color difference is present. 
This can be minimized through the use of color-matching filters. 
This leaves, then, only the inconvenience of taking a number of read- 
ings to average out errors of individual readings, and the fact that 
such instruments are less convenient to use. 

As a further possibility, the Sub-Committee has under considera- 
tion a photographic type of brightness meter. Such an instrument 
would function as a camera with a lens of fairly long focal length, and 
probably small aperture, that could be used to photograph the il- 
luminated screen from any seat position. Exposures would prob- 
ably be long perhaps in the range of ten to twenty seconds. 

The camera and film in itself could not be considered as an abso- 
lute brightness measuring device; an auxiliary piece of equipment to 
calibrate the camera and film at the time of its use would be essential. 
In its simplest form, this auxiliary device might consist of a box con- 
taining a lamp illuminating a series of targets to definite brightness 
levels. This panel box would have to be photographed either when 
the screen is photographed, or soon before or after, and both images 
would have to be developed together. The process of evaluating 
screen brightness and brightness distribution would involve simply 
comparison of target and screen image density distribution. 

There are certain difficulties in the design of such an apparatus. 
The color response of the film would have to be matched by filters to 
that of the eye. Precautions to avoid scattered light in the camera 
would have to be taken. Some convenient means of matching tar- 
get and screen-image densities would have to be devised. At best 
the apparatus involved would be cumbersome. 

On the other hand, such a device would result in a convenient form 
of permanent record of test results. In addition, the simplest form 
of operating instructions would suffice and it is beleived that the speci- 
fications of accuracy, angle, range, and price could be conveniently 
met. 



REPORT OF THE STANDARDS COMMITTEE* 

The two projects entitled "Designation of Direction of Winding of 
16-Mm Film Perforated along One Edge" and "Edge-Numbering 
Interval for 16-Mm Motion Picture Film" which were stated in our 
report at the Rochester Convention as being considered as SMPE 
Recommended Practices have now been approved, and have appeared 
in the November issue of the JOURNAL. 

In considering the "Designation of Direction of Winding of 16- 
Mm Film Perforated along One Edge" it was evident that there had 
been some divergence of practice among various companies in des- 
ignating the direction of winding 16-mm film that would naturally 
lead to some confusion among those who were handling film from a 
number of different companies. The designation that has been 
adopted as an SMPE Recommended Practice is as follows : 

When a roll of 16-mm film, perforated along one edge, is held so that the outside 
end of the film leaves the roll at the top and toward the right, winding A shall 
have the perforations on the edge of the film toward the observer; and winding 
B shall have the perforations on the "edge away from the observer. In both cases 
the emulsion surface shall face inward on the roll. This is illustrated by the 
following sketch. 




Winding A Winding B 

Emulsion side in Emulsion side in 



This sketch shows film wound on film cores. When the film is wound on a 
reel having a square hole on one side and a round hole on the other, the square 
hole shall be understood to be on the side away from the observer. 

* Presented at the 1941 Fall Meeting at New York, N. Y.; received October 
20, 1941. 

87 



88 STANDARDS COMMITTEE REPORT 

This definition has already been adopted by a number of the large 
film manufacturing companies. 

There were likewise a number of different systems proposed and 
tried for edge-numbering 16-mm motion picture film. These in- 
cluded placing numbers on the film at 16-frame intervals correspond- 
ing to one-foot intervals on 35-mm film, numbers at one-foot inter- 
vals on 16-mm film, and numbers at intervals corresponding to sec- 
onds of screen time. After considerable study and discussion with 
various companies of the industry the following SMPE Recom- 
mended Practice was adopted : 

If 16-mm film is edge-numbered, the interval between consecutive footage 
numbers shall be 40 frames. 

This seemed to give the best solution and also assumes that not 
all 16-mm film is edge-numbered. It does likewise give the advan- 
tage of not having the numbers so close together as would be the case 
if they were placed at 16-frame intervals. 

A project for revising the standards for 16-mm sound-track and 
scanning area is being considered by the Committee at the present 
time. Work on the Glossary is progressing and several lists of words 
have been checked over by the Committee. The question of 
sprockets has been reviewed and a resum6 of the situation on 16-mm 
sprockets is being compiled. A sub-committee is actively engaged 
in seeing what can be agreed upon for consideration as Recommended 
Practice for 35-mm sprockets specifications. 

A number of committees of the Society, such as the Sound Com- 
mittee and the Non-Theatrical Committee, are working on projects 
that have been referred to them by the Standards Committee. 

D. B. JOY, Chairman 

P. H. ARNOLD A. F. EDOUART R. MORRIS 

H. BAMFORD J. L. FORREST WM. H. OFFENHAUSER, JR. 

C. N. BATSEL G. FRIEDL, JR. G. F. RACKETT 

M. C. BATSEL A. N. GOLDSMITH W. B. RAYTON 

F. T. BOWDITCH H. GRIFFIN E. C. RICHARDSON 

M. R. BOYER A. C. HARDY H. RUBIN 

F. E. CARLSON P. J. LARSEN O. SANDVIK 

T. H. CARPENTER C. L. LOOTENS R. E. SHELBY 

E. K. CARVER J. A. MAURER J. L. SPENCE 

H. B. CUTHBERTSON G. S. MITCHELL E. W. TEMPLIN 

L. W. DAVEE K. F. MORGAN H. E. WHITE 

J. A. DUBRAY 



SOME EQUIPMENT PROBLEMS OF THE DIRECT 16-MM 

PRODUCER* 



LLOYD THOMPSON** 



Summary. The production of industrial films by the direct 16-mm method is 
now definitely out of the experimental stage. As more industrial work is done by this 
method there is an increasing demand for more and better 16-mm equipment suitable 
for professional use. Such equipment can be developed successfully only after the 
professional user has found by actual experience what he needs and wants. 

A number of 16-mm professionals were asked for suggestions as to what is needed. 
These suggestions combined with the author's own ideas gained over a period of ten 
years in the professional 16-mm field form the basis of this paper. Some of the ideas 
presented could be acted upon immediately; some can not be put into practice until 
the demand for 16-mm service becomes even greater. 

A leading motion picture technical journal recently said in its 
columns, "16-mm commercial filming has long since outgrown the 
experimental stage and become a legitimate and highly specialized 
field of professional cinematography. The technicians in this field 
stand definitely apart both from the 16-mm amateur and from the 
35-mm professional fellow." 

Today there are several organizations within this country devoted 
exclusively to producing business and educational films by the direct 
16-mm method. While it is true that a great deal of splendid equip- 
ment has been built for the 16-mm users during the past few years, 
there is still a considerable amount of equipment which the com- 
mercial 16-mm producer would like to see on the market. I believe 
that it is realized by most of these producers that some of the things 
they want will not be placed on the market for some time because 
there is not enough of a demand to warrant any manufacturer placing 
it in production. That, however, does not keep the producer from 
wanting it. 

In an attempt to find out what the 16-mm producer would like to 
have, the writer recently sent a questionnaire to a number of men 

* Presented at the 1941 Fall Meeting at New York, N. Y. ; received September 
10, 1941. 
** The Calvin Company, Kansas City, Mo. 

89 



90 L. THOMPSON \j. s. M. P. E. 

throughout the country who were known to be doing direct 16-mm 
work. It is true that the questionnaire did not reach as many as it 
might because some of the 16-mm producers have not made them- 
selves well enough known. However, it is believed that the most im- 
portant of these producers were reached and it is also believed that 
the most important of them answered the questionnaire. The ques- 
tionnaire was sent also to others directly interested in direct 16-mm 
production. It is hoped that this report will help to interest certain 
equipment manufacturers in placing on the market certain of this 
equipment wanted by 16-mm film users. 

FILM 

In making a survey of this type it was natural to begin by asking 
what improvements were most wanted in 16-mm raw film for pic- 
tures, for sound, and for duplicating. It was quite interesting to 
note that several asked for a 16-mm negative film with extremely fine 
grain for making original films. This was especially interesting 
since none of these people followed up their request under the ques- 
tion about laboratory services. In order to use such a negative film, 
successfully it would be necessary to solve a number of problems con- 
nected with using negative film in the 16-mm size. Since it is a 
poor policy to figure on editing an original negative and making all 
the release prints from the original negative, it becomes necessary to 
make a master positive from the original negative, and then make a 
duplicate negative from that and make release prints on it. In order 
to do this, it is almost necessary that better cameras and printers be 
devised in order to keep the picture steady. Better laboratory ser- 
vice must be made generally available if a film is to be put through 
all these processes and still retain any quality, such as is done in 
Hollywood. 

A number of direct 16-mm productions are now being used with 
all synchronous sound. This has brought up the need for an edge- 
numbered 16-mm film for original shooting, a problem that has been 
before the Standards Committee of the Society and has had consider- 
able discussion. Unfortunately it is impossible to get any except a 
few emulsions with the edge numbering. There seems to be little 
agreement as to which films shall be numbered and which shall not 
be numbered. At least some of the producers feel that all film to be 
used for original shooting should be numbered. If so, it means that 
all black-and-white reversal film that is sold should be edge-numbered. 



Jan., 1942] EQUIPMENT PROBLEMS OF 16-MM PRODUCER 91 

It means that all kodachrome film or other color-film should also be 
edge-numbered. The producer of direct 16-mm pictures is not par- 
ticularly interested in whether duplicating films or positive films are 
numbered, although there are a number of film libraries and other 
organizations who are interested in having such film edge-numbered. 
It is argued by the manufacturers that it will be impracticable to 
edge-number all 16-mm film. It is argued by the producers that it 
would be useless to number some of the films and not all of them. 
The reason is quite simple: One of the things that makes direct 16- 
mm production so versatile is that 16-mm film can be purchased in 
any part of the country. Therefore, a producer making a picture 
at anytime in any part of the country can obtain his film at almost 
any photographic store if he finds that he needs more than he has 
taken along for the job. Many times the producer finds that it is 
advantageous to have someone else in some other part of the country 
shoot certain scenes and send them to him. This keeps production 
costs low, and since reversal film is so well standardized by the labora- 
tories of the film manufacturers, he can do so and be assured of get- 
ting good quality. Many times it is desirable to use several different 
cameras in photographing certain pictures, and at times it is desirable 
to use cameras of different types such as the camera that is loaded 
with a magazine. 

If such a picture is to be edited by means of a work print and edge 
numbers are to be used in the editing, it is almost necessary that all 
the originals be edge-numbered or the system will be quite useless. If 
the films have been obtained at different sources or if the pictures 
have been taken by various photographers in different parts of the 
country, it is almost necessary that all the film be edge-numbered, 
otherwise a number of shots will probably be made on film that is 
not edge-numbered. The writer believes that the day will come when 
all 16-mm film will be edge-numbered, and at the present time it 
should be possible to obtain on special order edge-numbered film from 
any film manufacturer. 

It was to be expected that a number of persons would ask for im- 
provements in color-film and a method of obtaining cheaper color- 
prints. There were some suggestions made in regard to improving 
color-film but only one request for cheaper color-prints. (Since that 
time there has been a reduction in the price of color-prints in quan- 
tity.) The improvement that seems to be most wanted in 16-mm 
color-film is an indoor or Mazda-light type of film with a speed com- 



92 L. THOMPSON [J. s. M. P. E. 

paring with that of the fast black-and-white films now available. 
That is, undoubtedly, a large order but, nevertheless, is something 
that is wanted. A better treatment for the sound-tracks on color- 
film would also be desirable. 

Several producers expressed an interest in obtaining better quality 
in release prints; some of the features wanted are sharper prints, 
better contrast control, finer grain, and better quality in halftones. 
In the opinion of the writer, a number of these things are laboratory 
problems instead of improvements in raw film stock. If the proper 
film stock is used for shooting the original picture, and if it is photo- 
graphed, lighted, and processed with the thought in mind that it is to 
be used for making release prints, a great number of these things can 
be controlled quite satisfactorily, provided the laboratory uses the 
proper control instruments. A special color-film to be used only for 
duplicating has recently been placed on the market that will give 
even better color prints than we have been able to obtain in the past. 
Improved printing and improved processing technics will undoubtedly 
give even better results. 

EDITING EQUIPMENT 

Probably the thing that is lacked most in 16-mm equipment for 
industrial purposes is editing equipment. All the editing equipment 
that is available seems to be designed to sell at a price for the amateur 
users of film, or has been designed or remade from existing 35-mm 
equipment. Neither type of equipment is entirely suitable for pro- 
fessional 16-mm use. A truly professional 16-mm splicer should be 
as automatic in operation as possible. It should be accurately made 
and should be positive in its operations. It should make a straight 
splice. If such a machine could be sold at a fairly moderate price, a 
large number of them could be sold. If it can not be made at a mod- 
erate price, the sale will probably be somewhat limited, but there are 
still a number of persons who are trying to find such a splicer at most 
any price. If the splicer is to be used in editing 16-mm sound-films 
it should be kept as small as possible. One of the main difficulties 
with 35-mm equipment that has been converted for use with 16-mm 
film is that the equipment is so large and bulky and awkward that 
the editor has a great deal of difficulty in handling it. 

Up to the present time 16-mm film is edited somewhat differently 
from 35-mm film. In direct 16-mm practice the original film itself 
is many times edited. The person doing the editing does his own 



Jan., 1942] EQUIPMENT PROBLEMS OF 16-MM PRODUCER 93 

cutting and splicing, all at the same time. It is seldom that the di- 
rect 16-mm editor cuts up a work print, splices it together hurriedly 
with tape, and then has someone go through it and make the splices. 
More often he is likely to work with the original film, making his cuts 
and doing his splices all at the same time. In such a case it is neces- 
sary that he be surrounded with the proper rewinds, synchronizers, 
footage counters, sound take-offs, picture viewer, mechanism for 
projecting sound and picture, and the splicer. Unless the equipment 
is small and easy to handle it will be very inconvenient for him to 
work. All the people who answered the question on editing equip- 
ment in the questionnaire said that they were using at least some 
special equipment they had built for editing purposes. A few were 
using moviola machines. Others were using modified moviolas, and 
still others were using special equipment they had built themselves. 
It would be expected that some of the special equipment had been 
built and was being used because it was cheaper than buying a regular 
moviola machine. However, that was not always the case, because 
some of the specially built equipment was considerably more costly 
than a moviola. There were very few definite suggestions as to how 
16-mm editing equipment should be made. The author does not 
have a great many definite ideas as to how such a machine should be 
made. The equipment used for editing at The Calvin Company has 
been built up from a Bell & Howell projector-head and a Victor Ani- 
matograph sound-head. This combination was used because it was 
found that the Bell & Howell head could be attached to the Victor 
sound-head in such a way that the film containing the picture and the 
film containing the sound could be very easily threaded up together 
to run both the sound-track and picture at the same time. It was 
found also that the drum on the Victor sound projector could be very 
well used as a machine for picking up separate words from the sound- 
track. A motion type viewer rebuilt to provide more positive ac- 
tion and avoid danger from film scratches is used for checking indivi- 
dual frames of the picture. A four-sprocket synchronizer and a pair 
of rewinds that will handle four reels at one time are also used, and a 
number of small rewinds for handling short rolls of film. It is be- 
lieved that this equipment would form a good basis for building an 
editing machine to be used similarly to a moviola. Such a machine 
would, of course, need a number of refinements, and it would un- 
doubtedly be quite costly to build because it should be built very 
well and very accurately to protect the original film that must go 



94 L. THOMPSON [j. s. M. p. E. 

through a 16-mm editing machine. While such a machine is cer- 
tainly desirable from the standpoint of the 16-mm producer, this is 
probably one of the things that will not be available for some time. 
The writer recently saw plans for a very simple editing machine that 
seems to show great promise. There is now one under construction 
and it should be ready for trial before too long. 

CAMERAS 

Practically everyone who was asked expressed a desire for a truly 
professional 16-mm camera. Several of those answering the ques- 
tionnaire said that they thought the Berndt-Maurer camera would 
fill the bill and others expressed a desire for certain changes in it be- 
fore accepting it as the professional 16-mm camera they would like to 
have. 

Many of the producers of 16-mm films are using standard cameras 
specially adapted to suit their work. Many of the cameras have had 
special gadgets added to them for special purposes. Some are driven 
by synchronous motors obtainable on the open market, and others 
are driven by specially built synchronous motors. A number of them 
have been blimped for sound, and in this case the blimps have all been 
made by the users. Recently a blimp for the Kodak special has been 
placed on the market. We at The Calvin Company built our own 
blimp several years ago. It was entirely adequate for the sound-re- 
cording equipment we were using at that time. However, we in- 
stalled improved recording equipment and amplifiers some time ago 
and the blimp we were using was then found unsatisfactory. We 
constructed another blimp, of different and heavier construction, 
which has been doing a satisfactory job. However, it is slightly 
harder to handle, and we should like to have a lighter and more ef- 
ficient one, or, better still, we should like to have a 16-mm camera 
for sound shooting that does not require a blimp. At the present we 
feel that this is the only serious defect in the professional camera be- 
ing offered. 

Many of the pictures we are now making for industrial use are 
synchronized dramatic shows, and it is necessary that a quiet camera 
be used in photographing them. Over a period of years we have 
tried practically all the 16-mm cameras on the market, and 
during that time we have come to rather definite conclusions as to 
what we should like in a camera. First, a professional 16-mm camera 
should be silent enough to be used on the set for photographing syn- 



Jan., 1942] EQUIPMENT PROBLEMS OF 16 MM PRODUCER 95 

chronous sound pictures without a blimp. If it is not possible to 
build such a camera, we feel that some convenient and easy method 
of blimping it should be worked out so that the camera will not be- 
come too heavy and difficult to operate. We realize that this is 
quite a large order but we do not feel that it is impossible. 

The camera should have registration pins, and the pull-down claw 
should be as close to the aperture as possible so that the frame-line 
will remain steady and centered at all times. The camera should 
have side-guides to prevent the film from weaving. The shutter 
should not be connected to the claw because such an arrangement will 
cause it to run unevenly and produce flicker in the pictures; or some 
method of eliminating the flicker should be used. The camera 
should be driven by some type of synchronous motor, which should 
be connected to the shutter in such a way that the film flow through 
the camera will be perfectly even and the exposures quite uniform 
so as to prevent flicker. We have always found that it is easier to 
get smooth, evenly exposed pictures with a spring-driven camera 
than with a motor-driven camera. As a matter of fact, we have had 
a great deal of difficulty with motor drives on cameras. 

The film-gate of a professional camera should be constructed in 
such a way that it is impossible to scratch film. Some cameras on 
the market hold the film perfectly flat in the camera, but in doing so 
they quite often scratch the film and produce static. When such a 
condition exists in a camera we have found that it is impossible to 
eliminate it completely even though the film-gate is cleaned thor- 
oughly after every 100-ft roll of film. On the other hand, there are 
some cameras on the market that do not give this trouble, but in 
these cameras the film will not stay in its proper plane at all times 
and the pictures are likely to be very much out of focus. The manu- 
facturers are well aware of this, and during the past year have done a 
great deal to eliminate the difficulty. For the past nine months we 
have not had any difficulty of this particular sort, but in building a 
professional camera this is certainly a pitfall that should be avoided 
by the manufacturer. 

As far as we are concerned, such a sound camera can be of the sim- 
plest type. It can run at one synchronous speed of 24 frames a 
second. It need not have an automatic fade or dissolving shutter 
because we feel that all such effects can and should be made in the 
laboratory and not on the set. 

The camera should have a view-finder that is really accurate and 



96 L. THOMPSON [J. S. M. p. E. 

easily used. It is not necessary that one be able to record sound with 
the camera because we believe that all sound made in the studio 
should be made by the double system. It is not urgent that the 
camera have a lens turret if such a turret would cause any special 
difficulty in manufacture or would raise the price considerably. On 
the other hand, there are times when a turret can be used. The 
camera should have an accurate method of focusing on the film. In 
short, we want a quiet-operating, synchronous-speed camera that 
will produce the sharpest, steadiest picture possible. 

These are some of our ideas as to what we should like to have in a 
16-mm professional camera. They do not necessarily agree with 
some of the views expressed by others. We believe that most pro- 
ducers want a variable shutter on the camera. They have expressed 
a desire that the shutter be made automatic, for automatic fades 
and dissolves, and some have expressed a desire for one that will 
make wipe-offs. A number have expressed the opinion that a 16- 
mm professional camera should be able to record sound on the film by 
the single-system method. It goes without saying that everyone 
wants such a camera to handle 400 feet of film or more. 

There have been numerous rumors during the past several years 
about companies who were going to make 16-mm professional cameras 
that would soon be on the market. So far most of these rumors have 
been pretty consistently denied, although several companies have 
admitted that they have been working on such a camera. The de- 
tails of the cameras have been kept secret by the manufacturers. If 
any manufacturer would care to take us into his confidence, we should 
like very much to see what he had in mind and help him make a 
practical test of the camera. 

RECORDERS 

No suggestions were received for improving direct 16-mm recorders. 
One person answering the questionnaire said that 16-mm sound 
should be more faithful and more realistic and should have less 
boominess. However, this person said that this characteristic was 
not limited to sound recorded by the direct 16-mm method but was 
present in nearly all 16-mm film no matter what method was used in 
producing them. Another person answering the questionnaire said 
that he felt that 16-mm sound could be made better but that the next 
improvements were going to come in film recording stock. It is 
certainly true that the new Agfa high resolving sound recording stock 



Jan., 1942] EQUIPMENT PROBLEMS OF 16-MM PRODUCER 97 

for 16-mm has shown that improvements in film stock will improve 
the results in recording. We are of the opinion that a great many 
persons who are doing direct 16-mm recording are doing their re- 
cording in rooms that have not been correctly treated for sound-re- 
cording purposes. Much of the direct 16-mm sound that is being 
recorded is being made by talent that is untrained and unrehearsed. 
However, we must say that even the talent is continually becoming 
better. Voice-recording machines are becoming quite popular and 
common. As a result of this, combined with radio work and pic- 
ture work, there is continually more talent available and much of this 
talent works very hard to give better performances. This improve- 
ment, together with the best 16-mm recording equipment, develop- 
ing, and printing, and an improved technic of recording, will naturally 
be reflected in future pictures. 

LABORATORY SERVICES 

After the picture and the sound have been put on the film, there is 
then the problem of properly developing and printing it. Good 
laboratory service for 16-mm film has been very slow in coming. It is 
true that the authorized processing stations of the film manufacturers 
have given fairly good and consistent service on reversal film. They 
have probably been the most consistent in their work of any of the 
16-mm laboratories, but even here there are times when processing 
standards could be held to closer limits. Even though the service 
from these laboratories has been good, that still does not answer the 
problems involved in getting release prints. Very few of these labora- 
tories will attempt to develop sound-tracks made on 16-mm sound- 
recording film. As a result, the producer of 16-mm sound pictures 
has been forced to look elsewhere for his laboratory services in de- 
veloping his sound-tracks and the making of dupe negatives from 
original reversal film. The answers received on the questionnaire to 
the question about laboratory services were in one way rather dis- 
appointing and in another way they were rather flattering. Most 
of those who replied to the questionnaire stated that they were per- 
fectly satisfied with the laboratory services they are now getting. 
This, of course, is very nice for the laboratories to hear, but it was 
hoped that a number of good suggestions would be made for im- 
proving laboratory service. Nearly all those answering the ques- 
tionnaire stated that they would like to have optical effects available 
for 16-mm work. Some said that they thought this was too much of a 



98 L. THOMPSON u. S. M. P. E. 

problem to present to the laboratories at the present time. Others 
knew about the optical effects that we offer in our laboratories and 
were very well satisfied with the results they could get by this method. 
Some indicated that it would be desirable if some method could be 
worked out so that these effects could be used at less expense. One 
answer to the laboratory question was particularly interesting. It 
came from a man who has had a great deal of experience in develop- 
ing 16-mm equipment and who has had also a considerable amount of 
experience in laboratory work. I knew the man seven or eight years 
ago, at which time he told me that 16-mm would never be successful 
until the laboratories learned how to control and develop the film 
correctly. Today his suggestion as to the most needed laboratory 
service is, "Much better work." Although we operate a laboratory, 
we are inclined to agree with what he says. 

The duplication of 16-mm color-film was a distinct achievement. 
It would be highly desirable if some method could be found for mak- 
ing color-prints by an intermediate negative step so that the original 
film would not have to be used for making all release prints. We, of 
course, do not have any idea when such a process will be practicable. 
Special trick printers have been designed and built for making koda- 
chrome release prints containing optical effects, which will print such 
release prints with no more trouble than straight printing. 

On the other hand, we feel that there is still a great deal of work 
that can be done on 16-mm printers. Practically all the 16-mm 
printers that are really doing the best job of printing have been de- 
signed and built specially for the laboratories owning them. Anyone 
desiring to purchase such a printer on the open market will run into 
many difficulties and become very discouraged before he finds what 
he wants. There is probably not a great deal that can be done about 
it, because there is not a large market for such equipment, unless we 
consider the small laboratories who can not afford to pay very much 
for it. Such a condition, however, gives 16-mm a bad reputation 
because there will always be a large number of persons who will try 
to make prints on equipment that is not adequate, and many of those 
who see these prints will judge 16-mm by these results. 

Some time ago a paper was presented before the Society on the 
problem of blooping sound-tracks. At that time, there was nothing 
definite that could be offered for blooping 16-mm film, and as far as 
we know, nothing has been done about it since. Some sort of black 
blooping patch that could easily be applied would be the answer to 



Jan., 1942] EQUIPMENT PROBLEMS OF 16-MM PRODUCER 99 

most of the direct 16-mm blooping problems. There are some, of 
course, who will not agree, because they will want to work with 
negative sound-tracks on which a black blooping patch will not work. 
We believe it will answer many of the problems because a great deal 
of the direct 16-mm sound is being made for kodachrome printing, in 
which case either a direct positive track can be made and used or a 
positive track printed from the negative. In either case the black 
blooping patches can be used on the edited positive track used for 
printing. For the person who uses a negative track, some sort of 
positive should be used for editing the picture. The splices can then 
be blooped out with the black patch and a re-recording or even a 
dupe negative made from this edited positive track. The re-recorded 
track is better for release printing because it can be made all in one 
piece and of one density for printing. We understand that such 
blooping patches were used some time ago on 35-mm film, and on 
several occasions we tried to purchase some of them without any luck. 
There may be some reason why they are not successful, or it may be 
that we have not tried to buy them from the correct source. In any 
event, we should like to be enlightened on this point. We have always 
painted out all our splices, and we find that that works entirely 
satisfactorily provided some care is used in painting them out. How- 
ever, it is a rather slow process and, furthermore, we have had to de- 
velop our own painting-out fluid because the fluids ordinarily sold 
to be used with nitrate film do not seem to work successfully with 
safety film. It has been some time since we tried the black-out 
paint obtainable on the market but it is also my recollection that 
this material was slow in drying. That was not objectionable when 
only one or two splices were to be painted out, but it became quite a 
problem when a number of splices were involved. A paint-out fluid 
should dry very rapidly. It should cover well and should stay on 
safety film. It should be easy to apply and should not chip or crack. 

SOUND PROJECTORS 

The question dealing with 16-mm sound projectors was probably 
the most interesting question in the lot, probably because more per- 
sons have had experience with sound projectors than with some of 
the other equipment mentioned. One hundred per cent of those who 
answered the questions on projectors said they did not feel that any 
16-mm sound projector on the market at the present time was en- 
tirely satisfactory from all standpoints. The question was probably 



100 L. THOMPSON [j. s. M. p. E. 

unfair because no piece of machinery will ever be entirely satisfactory 
from all standpoints in the opinion of all people. One hundred per 
cent of the replies picked out one brand of projector and indicated 
that it probably was the best. However, on this same question 20 
per cent also included another projector that they thought was very 
satisfactory. Ten per cent picked a third projector as being satisfac- 
tory. One did not answer the question on projectors. 

Sixteen-mm sound projectors are today performing very satisfac- 
torily. If all the manufacturers of projectors could get together and 
combine all the good ideas that have been used on the various ma- 
chines, one would be produced that would be almost perfect. How- 
ever, no one manufacturer has been able to do so, and as a result we 
still have certain objections to various machines. Some of the faults 
will be found on one machine and some on others. First, we should 
like to see a machine that will produce sharper pictures on the screen. 
Much 16-mm film has a tendency to be slightly curly, and since it 
does not lie flat on the gates at all times, the projected picture will 
often have fuzzy edges or a fuzzy center. If the film is in perfect 
condition most any of the projectors will throw a sharp picture on the 
screen, but there seem to be some machines that are much better in 
this respect than others. The speed and type of lens used will, of 
course, affect the situation. A very complete report was given in 
the July, 1941, issue of the JOURNAL by the Committee on non-the- 
atrical equipment. 

We should like better sound from nearly all the projectors. Am- 
plifiers should be better standardized. It seems that there are no 
two projectors of the same make or of different makes that will play a 
sound-track and make it sound the same. One can start off across 
the country with a demonstration reel that is felt to be pretty good 
and before the trip is half over he will begin to wonder how he ever 
thought it was a good demonstration reel. Of course, this may be 
the result of not keeping the projectors in proper operating condition. 
This is not always so, however, because one can take a batch of new 
projectors coming from any one of the manufacturers with whom we 
have had experience and almost always find something that is not 
adjusted as it should be. Of course, the manufacturer might argue 
that the dealer or producer selling the machine should see that the 
machine is in proper operating condition before it is released to the 
customer. We agree with that, but we think also that the manu- 
facturer should do some training work along this line in order that 



Jan., 1942] EQUIPMENT PROBLEMS OF 16-MM PRODUCER 101 

the producer and other distributors of sound projectors will be as 
familiar with the operation of the machine as an automobile dealer's 
mechanic is with the automobile he sells. 

One of the biggest improvements that could be made by all pro- 
jector manufacturers is to increase the power of their amplifiers. 
This might, of course, increase the cost and weight of the projector 
in some instances, but we feel that the increased operating efficiency 
would make it well worth while. Nearly all the lower-priced pro- 
jectors give trouble in this respect. Many times sound-prints are 
blamed for poor sound quality when there is nothing wrong except 
that the amplifier is being pushed too hard. The use of kodachrome 
sound-prints has made it even more desirable to have an amplifier 
of sufficient power. 

We feel that all projectors should include a method of focusing the 
sound-track light-beam for prints of different types. Such devices 
are available on special order, but should be made standard equip- 
ment. One projector has sound-track focusing as standard as a 
standard feature; the sound focusing is done exactly as in the picture 
focusing, and it works very well. 

Some projectors have trouble with their take-ups and rewinds, 
others are perfect in this respect. Some projectors still have bad 
"wows" and others give no trouble at all in this respect although there 
are some that have some sprocket flutter. 

Practically every manufacturer of sound projectors advertises 
the fact that his projector will not scratch film. We have never 
found a projector which would not scratch film. We grant that it 
is quite an engineering problem to produce a machine that will not, 
under any condition, scratch film, but we still have hopes. Some 
projectors have gates that are difficult to clean, and others are easy 
to clean. Some projectors are difficult to thread and others are fairly 
easy to thread. All projectors should be easy to thread. All 16-mm 
sound projectors should be quieter in operation. As better sound 
quality is put on 16-mm sound-tracks quietness will become more 
and more important. 

DISCUSSION 

MR. HYNDMAN: Perhaps the law of supply and demand is the real factor 
facing the 16-mm motion picture producer when he considers the availability of 
precision equipment to do his work. Manufacturers of 16-mm cameras, sound 
recorders, printers, projectors, and other miscellaneous equipment undoubtedly 



102 L. THOMPSON 

would be glad to supply high-precision equipment if there were a sufficient number 
of prospective consumers willing to pay the price. 

Most of the present equipment meets a volume market demand at low cost 
per unit to the consumer. Unfortunately, many users of 16-mm equipment fail 
to appreciate that the construction of high-precision 16-mm cameras, printers, 
and projectors is just as costly as the construction of similar 35-mm equipment. 

It has not been practicable, apparently, for the 16-mm producer to assume 
the obligation of the cost of design, development, and construction of high- 
precision equipment because the unit cost would not justify the present returns. 
When there is sufficient demand for this type of equipment, then the manu- 
facturers, very likely, will supply it at a figure that would be practical to the 16- 
mm producer. 



CURRENT LITERATURE OF INTEREST TO THE MOTION PICTURE 

ENGINEER 



The editors present for convenient reference a list of articles dealing with subjects 
cognate to motion picture engineering published in a number of selected journals. 
Photostatic copies may be obtained from the Library of Congress, Washington, D. C. t 
or from the New York Public Library, New York, N. Y. Micro copies of articles 
in magazines that are available may be obtained from the Bibliofilm Service, Depart- 
ment of Agriculture, Washington, D. C., at prevailing rates. 



American Cinematographer 

22 (November, 1941), No. 11 

Gasparcolor Comes to Hollywood (pp. 510-511) 

Proving the New Norwood Exposure Meter on Produc- 
tion (pp. 512-513), 539-540) 

Corrective Make-Up Can Help the Cinematographer 
(pp. 514, 540) 

Blue Windows (p. 517) 

An Amateur Tries 16-Mm Sound-on-Film Record- 
ing (pp. 523, 542-543) 

Educational Screen 

20 (November, 1941), No. 9 
Motion Pictures Not for Theaters (pp. 383-385), Pt. 31 

International Projectionist 

16 (September, 1941), No. 9 
Advance Signs of Reproducer Trouble (pp. 7-8, 11), Pt. 

II 

The RCA 866A/866 M.-V. Rectifier Tube (p. 11) 
Fantasound: A Technologic Epoch (pp. 12-14), Pt. II 

Theater Television: Some Technical and Economic 
Aspects (pp. 15-16, 21-26) 

Motion Picture Herald (Better Theaters Section) 
145 (November 15, 1941), No. 7 
Modern Wiring of Projection Rooms (pp. 33-34) 



A. WYCKOFF 
W. STULL 

J. DAWN 
R. METTY 

K. O. HEZZELWOOD 



A. E. KROWS 



L. CHADBOURNE 

W. E. GARITY AND 
J. N. A. HAWKINS 

A. HYNE 



103 



SOCIETY ANNOUCEMENTS 
FIFTY-FIRST SEMI-ANNUAL CONVENTION 

HOLLYWOOD, CALIF. 
MAY 4-8, 1942 

The 1942 Spring Convention of the Society will be held at Hollywood, Calif., 
with headquarters at the Hollywood-Roosevelt Hotel. The dates of the Con- 
vention closely precede those of the Convention of the National Variety Clubs of 
America, to be held at Los Angeles, May 14th, 15th, and 16th. 

Details of the Convention will be announced in the next issue of the JOURNAL. 
In the meantime the Papers Committee and the Convention Arrangements Com- 
mittee are proceeding with plans to make the convention an outstanding one 
despite the difficulties of the times. It is felt that every effort should be made 
during the emergency to keep up the technical activities of the Society and the 
industry, and members are urged to make all possible efforts to attend the Con- 
vention. 

Special evening sessions will be provided, so as to make it possible for those 
engaged hi the studios during the daytime to attend the technical meetings. 
Those contemplating the preparation of papers for presentation at the meetings 
should communicate with the General Office of the Society as promptly as possible. 

ATLANTIC COAST SECTION 

The meeting of the Section held at the Hotel Pennsylvania on December llth 
was devoted to the documentary film. Mr. Irving Lerner, Chairman of the 
Education Committee of the Association of Documentary Film Producers, 
Inc., and Mr. Richard Griffith of the Museum of Modern Art Film Library 
discussed the problems of producing documentaries and traced their develop- 
ment during the past number of years. A film A Place to Live was shown as 
a typical example of the documentary film. 

PACIFIC COAST SECTION 

A special showing of several subjects produced by major studios for the U. S. 
Signal Corps was held at the Paramount Studio Theater on November 27th. 
The presentation was supplemented by a paper on U. S. Signal Corps production 
activities by Major Charles S. Stodter, Signal Corps Liaison Officer. 

The subjects shown included Military Courtesy and Customs of Service, Basic 
Principles of Skiing, The 60-Mm Mortar Mechanical Training, The Anti- Aircraft 
Searchlight Battery. Colonel Nathan Levinson, U. S. Signal Corps Reserve, was 
Chairman of the meeting. 



104 



ADMISSIONS COMMITTEE 



At a recent meeting of the Admissions Committee, the following applicants for 
membership were admitted into the Society in the Associate grade: 

BENHAM, H. J. LYNCH, W. J. 

RCA Manufacturing Co., Inc., Hotel Victoria, 

Camden, N. J. 51st St. & 7th Ave., 

BRANCH, RAY New York, N. Y. 

Strand Theater, MARANO, Luiz 

Hastings, Mich. 42, Rua do Passeio, 

BROWN, C. R. Rio de Janeiro, Brazil 

General Electric Co., ROBIN, J. E. 

1 River Road, 330 West 42nd St., 

Schenectady, N. Y. New York, N. Y. 

ENKE, E. E. ROTTELL, I. J. 

2440 N. Sawyer Ave., 16128 Wisconsin Ave., 

Chicago, 111. Detroit, Mich. 

GATY, C. B. TOWLE, P. A. 

Fairchild Aviation Corp., Box 263, 

88-06 Van Wyck Blvd., Marshall, Mich. 

Jamaica, N. Y. UNGERER, J. S. F. 

GOLDMAN, R. Union Education Dept., 

120 Broadway, Union Government, 

New York, N. Y. Pretoria, South Africa 
VICTOR, W. S. 
376 Harvard St., 
Rochester, N. Y. 



In addition, the following applicant has been admitted to the Active grade: 

MILLS, B. E. 

Mills Novelty Co., 

4100 W. Fullerton Ave., 
Chicago, 111. 



105 



BACK NUMBERS OF THE TRANSACTIONS AND JOURNALS 

Prior to January, 1930, the Transactions of the Society were published quar- 
terly. A limited number of these Transactions are still available and will be 
sold at the prices listed below. Those who wish to avail themselves of the op- 
portunity of acquiring these back numbers should do so quickly, as the supply 
will soon be exhausted, especially of the earlier numbers. It will be impossible 
to secure them later on as they will not be reprinted. 



1924 



1925 



No. 

19 

20 
21 
22 
23 
24 



Price 
$1.25 
1.25 
1.25 
125 
1.25 
1.25 



1926 



1927 



No. 

25 

26 

27 

28 

29 

32 



Price 
$1.25 
1.25 
1.25 
1.25 
1.25 
1.25 



1928 



1929 



No. 

33 

34 
35 
36 
37 
38 



Price 
$2.50 
2.50 
2.50 
2.50 
3.00 
3.00 



Beginning with the January, 1930, issue, the JOURNAL of the Society has been 
issued monthly, in two volumes per year, of six issues each. Back numbers of 
all issues are available at the price of $1.00 each, a complete yearly issue totalling 
$12.00. Single copies of the current issue may be obtained for $1.00 each. 
Orders for back numbers of Transactions and JOURNALS should be placed through 
the General Office of the Society and should be accompanied by check or money- 
order. 



SOCIETY SUPPLIES 

The following are available from the General Office of the Society, at the prices 
noted. Orders should be accompanied by remittances. 

Aims and Accomplishments. An index of the Transactions from October, 
1916, to December, 1929, containing summaries of all articles, and author and 
classified indexes. One dollar each. 

Journal Index. An index of the JOURNAL from January, 1930, to December, 
1935, containing author and classified indexes. One dollar each. 

Motion Picture Standards. Reprints of the American Standards and Recom- 
mended Practices as published in the March, 1941, issue of the JOURNAL; 50 cents 
each. 

Membership Certificates. Engrossed, for framing, containing member's* name, 
grade of membership, and date of admission. One dollar each. 

Journal Binders. Black fabrikoid binders, lettered in gold, holding a year's 
issue of the JOURNAL. Two dollars each. Member's name and the volume 
number lettered in gold upon the backbone at an additional charge of fifty cents 
each. 

Test- Films. See advertisement in this issue of the JOURNAL. 



JOURNAL OF THE SOCIETY OF 
MOTION PICTURE ENGINEERS 

VOLUME XXXVIII FEBRUARY, 1942 

CONTENTS 

A Frequency-Modulated Control-Track for Movie- 
tone Prints 

J. G. FRAYNE AND F. P. HERRNFELD 111 

Design and Use of Noise-Reduction Bias Systems 

R. R. SCOVILLE AND W. L. Bell 125 

A Precision Direct-Reading Densitometer 

M. H. SWEET 148 

An Analysis of the Application of Fluorescent Lamps 
to Motion Picture Photography R. ROSENBERG 173 



Iodide Analysis in an MQ Developer 

R. M. EVANS, W. T. HANSON, JR., AND P. K. GLASOE 180 

Synthetic Aged Developers by Analysis 

R. M. EVANS, W. T. HANSON, JR., and P. K. GLASOE 188 

Current Literature 207 

1942 Spring Convention at Hollywood, Calif., May 
4th-8th 209 

Society Announcements 213 

(The Society is not responsible jor statements of authors.) 



JOURNAL OF THE SOCIETY OF 
MOTION PICTURE ENGINEERS 

SYLVAN HARRIS, EDITOR 

Board of Editors 

ARTHUR C. DOWNES, Chairman 

JOHN I. CRABTREE ALFRED N. GOLDSMITH EDWARD W. KELLOGG 

CLYDE R. KEITH ALAN M. GUNDELFINGER CARLETON R. SAWYER 

ARTHUR C. HARDY 

Officers of the Society 

^President: EMERY HUSE, 6706 Santa Monica Blvd., Hollywood, Calif. 
* Past-President: E. ALLAN WILLIFORD, 30 E. 42nd St., New York, 

N. Y. 
*Executive Vice-President: HERBERT GRIFFIN, 90 Gold St., New York, 

N. Y. 
^Engineering Vice-President: DONALD E. HYNDMAN, 

350 Madison Ave., New York. N. Y. 
^Editorial Vice-President: ARTHUR C. DOWNES, 

Box 6087, Cleveland, Ohio. 

* Financial Vice-President: ARTHURS. DICKINSON, 28 W. 44th St., New 

York, N. Y. 

* Convention Vice-President: WILLIAM C. KUNZMANN, 

Box 6087, Cleveland, Ohio 

*Secretary: PAUL J. LARSEN, 44 Beverly Rd., Summit, N. J. 
^Treasurer: GEORGE FRIEDL, JR., 90 Gold St., New York, N. Y. 

Governors 

*MAX C. BATSEL, 501 N. LaSalle St., Indianapolis, Ind. 
**FRANK E. CARLSON, Nela Park, Cleveland, Ohio. 

*JOHN G. FRAYNE, 6601 Romaine St., Hollywood, Calif. 

*ALFRED N. GOLDSMITH, 580 Fifth Ave., New York, N. Y. 
**EDWARD M. HONAN, 6601 Romaine St., Hollywood, Calif. 

*I. JACOBSEN, 177 N. State St., Chicago, 111. 
**JOHN A. MAURER, 117 E. 24th St., New York, N. Y. 

*LOREN L. RYDER, 5451 Marathon St., Hollywood, Calif. 

* Term expires December 31, 1942. 
** Term expires December 31, 1943. 



Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, included 
in their annual membership dues; single copies, $1.00. A discount on subscription or single 
copies of 15 per cent is allowed to accredited agencies. Order from the Society of Motion 
Picture Engineers, Inc., 20th and Northampton Sts., Easton, Pa., or Hotel Pennsylvania, New 
York.N.Y? 

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 

General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 

Entered as second-class matter January 15, 1930, at the Post Office at Easton, 

Pa., under the Act of March 3, 1879. Copyrighted, 1942, by the Society of Motion 

Picture Engineers, Inc. 



A FREQUENCY^MODULATED CONTROL-TRACK FOR 
MOVIETONE PRINTS* 



J. G. FRAYNE AND F. P. HERRNFELD** 

Summary. A 5-mil frequency-modulated track located between sound and pic- 
ture areas is proposed to control reproduction in the theater from one or more sound- 
tracks. A variation of approximately one octave in the control frequency provides a 
30-db change in volume range which may be used in part for volume expansion of loud 
sounds or as noise reduction for weak sounds. The control-track frequency is varied 
manually and recorded simultaneously with the sound-track in the dubbing operation, 
the gain of the monitoring channel being varied in accordance with the control fre- 
quency to produce automatically the enhanced volume range desired from the release 
print. The track is recorded in line with the standard sound-track and does not re- 
quire separate printing or reproducing apertures. It is scanned by a separate photo- 
sensitive surface, the output being converted from frequency to voltage variations by a 
frequency-discriminating network identical to that used in the monitoring channel. 
The output from the network, applied to the grid of a variable-gain amplifier in the 
sound channel, controls automatically the volume of the reproduced sound in accord- 
ance with that observed in the dubbing operation. 

It is well established that the sound-on-film medium that has 
been employed since the inception of sound pictures has a limited 
volume range which is incapable of reproducing in the theater with- 
out external aids the range of sound intensities picked up in the record- 
ing process. While the recent introduction of fine-grain film has in- 
creased the basic signal-to-noise ratio by several decibels the increased 
volume range thus obtained is still far short of meeting the volume 
range requirements of modern sound pictures. It is still necessary 
to expand the volume range by using devices such as noise reduction. 
In the variable-density method of recording both squeeze-track and 
printer-light variation are used to expand the volume range fur- 
ther. Experience has shown that by judicious use of these three 
technics approximately 20 db may be added to the range which is 
obtained from a sound record made without employing any of these 
devices. 

*Presented at the 1941 Fall Meeting at New York, N. Y.; received October 
20, 1941. 

** Electrical Research Products, Inc., Hollywood, Calif. 

Ill 



112 J. G. FRAYNE AND F. P. HERRNFELD [J. S. M. p. E. 

It has been found that even when these technics are employed the 
volume range is still insufficient for correct reproduction of extremely 
loud sounds often required to enhance the dramatic presentation of a 
sound picture, and the resort to overloading of the light- valve is only 
too common a practice in such situations. To remedy this situation 
the use of a control-track has been suggested which would produce an 
automatic change in gain in the reproducing sound system and which 
would make it possible to reproduce in the theater a range of sound 
volume comparable to that originally existing on the sound-recording 
stage. The use of such a control-track makes it possible to eliminate 
some of the technics enumerated above, especially squeeze-track and 
printer-light control, and thereby simplifies the operations necessary 
for producing the enhanced volume range in the theater. 

Proposals have been made from time to time for a control-track 
located on various parts of the movietone print, such as the area be- 
tween adjacent sprocket-holes, as suggested in British patent No. 
473,256 and U. S. patent No. 2,199,559, and currently employed in 
the Vitasound System. 1 It has also been suggested that the area 
outside the sprocket-holes on the sound-track side of the print be 
employed for control-track purposes. The control-track located in 
the sprocket-hole area is open to several objections. For example, 
the presence of 96 cycles and its numerous harmonics limits the use 
of this area to control frequencies well below 96 cps. This permits 
ordinary manual operation of the controls but prevents the use of 
any automatic control that might follow the sound envelope fre- 
quencies. It is difficult to superimpose multichannel control fre- 
quencies on the sprocket-hole track due to the narrow frequency range 
available below 96 cycles. It is also difficult to record and scan this 
track in line with the sound-track, thereby requiring separate record- 
ing, printing, and reproducing apertures. The location of the sound- 
track outside the sprocket-holes is also subject to several objections, 
such as the presence of footage marks and other printed informa- 
tion, the liability to wear and tear, and the accumulation of oil and 
other dirt which might interfere with the proper functioning of the 
track in this area. It requires also, in common with the control- 
track in the sprocket-hole areas, the addition of separate recording, 
printing, and scanning apertures. 

An examination of a movietone print shows that there is an un- 
used area 0.029 inch wide between the inner edge of a standard 76- 
mil sound-track and the outer edge of the picture frame. Allowing 



Feb., 1942] CONTROL-TRACK FOR MOVIETONE PRINTS 



113 



for an 84-mil scanning aperture in the theater reproducing equipment, 
which overlaps the 76-mil track by 4 mils on either side, there re- 
mains an effectively unused area of 0.025 inch between the scanning 
aperture and the picture. The control-track described in this paper 
is located in this area, having the dimensions shown in Fig. 1. The 
5-mil width of the control-track was selected only after tests had 
shown that the output from such a track was ample for all intended 




.076 



t.005"CON' 
SIGNAL T 



CONTROL TRACK 
TRACK 



.243" 



FIG. 1. The control-track. 



operations. The location of 16 mils from the sound-track and 8 mils 
from the picture area was chosen with due regard to established toler- 
ances in printing and reproducing machines. It will be noted that a 
scanning aperture 105 mils wide is required for proper scanning of the 
sound and control-tracks. It will also be noted that the sound-track 
is symmetrically placed between the narrow control-track and the 
sprocket-holes and is located in the standard position on the film, 



114 J. G. FRAYNE AND F. P. HERRNFELD [J. S. M. P. E. 

thereby permitting playing in a theater not equipped for control- 
track reproduction. 

Amplitude Modulated Control-Track. In the first attempt to use a 
control- track in this area, a single control-frequency was recorded, 
the amplitude being varied manually in the recording process in ac- 
cordance with the sound level desired from the print. Various single 
frequencies, ranging from 7000 cycles down to 1000 cycles, were em- 
ployed at different times, the use of any frequency within this range 
permitting so-called "fast" operation of the control-track. In re- 
production the 5-mil control-track was scanned by a separate photo- 
electric cell placed in the sound-head, and the output fed into a 
specially designed logarithmic amplifier. The output from this am- 
plifier was rectified, filtered, and the resulting voltage applied to a 
variable-gain stage of amplification in the signal channel. Since the 
signal-to-noise ratio of a 5-mil track is approximately 23 db lower than 
on a 76-mil track, it was necessary to pass the control-signal through 
a narrow band-pass filter to secure a sufficiently high signal-to-noise 
ratio. With a band-pass of = fc 250 cycles, a signal-to-noise ratio of 38 
db was obtained which was ample for a 30-db range of volume con- 
trol. 

The chief objection to the use of an amplitude-modulated control- 
track is that the amount of expansion is subject to variations in the 
output of the control-track photocell, which may be caused by fluc- 
tuations in sound-track density, reprbducing-lamp intensity, or photo- 
cell sensitivity. The amount of expansion is, of course, subject 
also to accumulation of dirt and scratches on the control-track 
which tend to vary the transmission of the track. The control 
frequency is also subject to modulation at a 24-cycle rate by the 
"burn-over" of the adjacent frame lines. While this effect can be 
eliminated by insertion of a suitable filter, the operating time must 
necessarily be limited to values greater than one twenty-fourth of a 
second. It also requires the construction of rather complicated con- 
trol amplifying equipment, the characteristics of which depend to 
some extent on the characteristics of the particular vacuum-tubes 
employed. 

Frequency Modulated Control Track. For these reasons it was 
decided to apply the principle of frequency-modulation to the con- 
trol-track. In this case the frequency to be recorded on the control- 
track is varied, the amplitude being kept constant, the changing 
frequency in turn producing the desired changes in loudness of the 



Feb., 1942] CONTROL-TRACK FOR MOVIETONE PRINTS 115 

reproduced sound. In the reproducing equipment, the frequency 
variations are converted to amplitude variations by the use of a suit- 
able discriminating network, the rectified output of which is again 
employed to change the bias in the variable-gain stage of amplifica- 
tion in the reproducing signal channel. Experience with this type of 
control-track has shown that it is not subject to any of the limitations 
previously found for the amplitude-modulated track and that it 
tends to be much more reliable under theater operating conditions. 

Method of Recording Control-Track. Since the control-track is in- 
tended for use on the release print, it is recorded during the dubbing 
operations on the release negative. An RA-1061 push-pull type 
light-valve is employed to record simultaneously both signal and 
control- tracks. The signal is applied to one pair of ribbons and is 
recorded as a standard 76-mil sound-track, while the second pair of 
ribbons with suitable masking in the pole-pieces is used to lay down 
the 5-mil control-track. The various individual tracks are mixed in 
the usual manner to maintain the proper balance between music, 
dialog, and sound effects, and ample modulations of the light-valve 
without overload should be maintained irrespective of the resulting 
sound volume. The enhanced sound volume, which is heard directly 
over the expanded PEC monitoring system, is controlled by varying 
the frequency impressed on the control-track. This frequency is 
determined by varying the resistance elements of a variable-frequency 
oscillator which is located in the mixing console. Provision is made 
for either direct monitoring of the unexpanded signal being recorded 
or of PEC monitoring of the expanded signal that will later be repro- 
duced from a control-track print. 

The control-track may be used either to enhance the volume of 
loud sounds which normally are compressed due to the limited volume 
range of the film medium, or may also be used at the other end of the 
sound-intensity scale to reduce the background noise of the film. 
Thus, instead of recording low-level passages, as is customary 
at a low per cent modulation of the light-valve, these passages may 
be recorded up to nearly top level, and the proper sound balance 
restored by using the control-track to reduce in proportion the gain 
of the reproducing channel. In practice it has been found that with 
a total of 30 db of volume-control range in the reproducing system, 
the top 20 db of this range may be successfully employed for expan- 
sion of loud musical passages and sound effects, normal dialog level 
being recorded at the 10-db expansion level. To permit increasing 



116 



J. G. FRAYNE AND F. P. HERRNFELD [J. S. M. p. E. 



the film modulation for low-level passages, the lower 10-db range of 
the control-track may be utilized. Thus, it seems feasible not only 
to expand the louder sounds by as much as 20 db but effectively to 



RE- RECORDER 


<^ 




- 




J 






>PEC. 
5-r 




DIRECT 


VC 


5>r 


AMPLIFIER 


NOISE 






UNIT 


RE-RECORDER 




I 


PEC C 




4 




AMPLIFIER 


VAR. GAIN 





OTHER 
RE-RECORDERS 




-f s 


AMPLIFIER 


?. L c 




SIGNAL LV 

OH 








CONTROL 
OSCILLATOR 


DISCRIMINA- 
TOR 





CONTROL LV 



FIG. 2. Re-recording and monitoring system. 

reduce background noise during low-level passages by as much as 10 
db by the automatic reduction in gain of the reproducing system by 
this amount through the operation of the control-track. 



4,000 



DC 3,000 



2,000 



10,000 20,000 30,000 40,000 50,000 

OHMS 
FIG. 3. Oscillator control curve. 



Monitoring and Reproducing Circuits. Since the circuits used for 
monitoring the expanded volume range in the re-recording operation 
are identical with those intended for reproducing the control-track 
film in the theater, it is only necessary to explain their operation in 
the re-recording process. A diagram of the re-recording and moni- 



Feb., 1942] CONTROL-TRACK FOR MOVIETONE PRINTS 



117 



toring layout is shown in Fig. 2. The outputs from the various re- 
recording machines are mixed as usual into a single sound channel 
shown in the drawing. The resulting signal frequencies are amplified 
and applied to one pair of ribbons in the RA-1061 valve in the cus- 
tomary manner. Simultaneously the control-frequency generated 
by a variable-frequency oscillator is applied to the second pair of 
ribbons of the light-valve. This oscillator may be any one of a 
variety of oscillators provided that the oscillator used is of such na- 
ture that simple controls may be utilized to vary the output of the 
oscillator from 2000 to 4000 cycles. This particular range of fre- 
quency was chosen for the experimental work but any other equiva- 
lent range in the sound-frequency spectrum may be equally well 
utilized. The relationship between frequency and control resis- 
tance of the particular oscillator employed is shown in Fig. 3. By 



PEC 


AMPLIFIER 





CONTROL VOLTAGE 
OUTPUT 



FIG. 4. Discriminator unit. 

means of the double PEC monitoring arrangement shown in Fig. 2, 
the control-track signal is picked up by the second photocell and fed 
to the discriminator unit shown in Fig. 4. 

The output of the control-track is first amplified and then trans- 
mitted through a high-pass filter to prevent all extraneous frequencies 
below the lower 2000-cycle limit from affecting the operation of the 
frequency-discriminating circuit. The control-signal is next passed 
through a limiting amplifier employing grid and plate saturation 
which has an output vs. input characteristic as shown in Fig. 5, this 
limiting action being necessary to insure that the voltage input to the 
bridge circuit is constant and independent of frequency. This 
bridge, which is of the type disclosed in U. S. patent No. 2,106,785, 
assigned to the Bell Telephone Laboratories, is balanced at 4000 
cycles and serves as a frequency discriminator to convert the fre- 



118 



J. G. FRAYNE AND F. P. HERRNFELD [J. S. M. P. E. 



quency variations to voltage variations. The insertion-loss char- 
acteristic is shown in Fig. 6, the loss at the balance frequency amount- 
ing to 67 db. This so reduces the output in the operating region 
that 45 db of gain is required to obtain the necessary voltage for ap- 



+10 



D 
Q. 

I -10 



-60 -40 -20 

INPUT IN DB 0=. 006 WATTS 

FIG. 5. Limiting amplifier characteristic. 



+ 20 



100 
80 
60 

20 

























\ 






















\ 






















\ 


x 






















>s 


^ 






















^ 


^^ 




2000 3000 4000=fr 



FIG. 6. Insertion-loss characteristic. 



plication to the control grids of the variable-mu tubes in the variable- 
gain stage in the signal channel. The output of the amplifier is 
rectified and then transmitted through a combined low-pass and R. 
C. filter to prevent noise and extraneous frequencies above the bal- 
ance point from being transmitted to the signal channel. The 



Feb., 1942] CONTROL-TRACK FOR MOVIETONE PRINTS 



119 



operating time of 34 milliseconds and release time of 150 milliseconds 
are determined by the particular filter circuits employed and are not 
at all critical since the circuit may be used with an operating speed 
up to 3 milliseconds if so desired. 

The output of the signal circuit is fed first into two stages of a re- 
sistance-coupled amplifier. The push-pull variable-gain stage follows 
the pre-amplifier. The gain of this stage is controlled by the biasing 
voltage applied to the variable-mu tubes from the control- track cir- 
cuit described above. This bias is fed through a balanced-bridge 



30 



20 



10 



10 15 20 25 

CONTROL VOLTAGE 



30 



FIG. 7. Variable-mu stage characteristic. 

circuit to the control-grids in order to eliminate any residual ripple 
which may have been transmitted through the filter in the discrimi- 
nator. In order to permit the use of this amplifier for the reproduc- 
tion of standard sound-films, provision is made to switch in a fixed 
bias in the variable-mu stage to replace the biasing voltage supplied 
from the control-track. The relationship between gain of the vari- 
able-mu stage and control voltage is shown in Fig. 7, where it will be 
noticed that 30 db is obtained from a bias voltage range of approxi- 
mately 24 volts. The relationship between gain of the variable-mu 
stage and frequency of the control-track is shown in Fig. 8, 



120 



J. G. FRAYNE AND F. P. HERRNFELD [J. S. M. P. E. 



Reproduction Equipment and Circuits. Since it has been pointed 
out that the circuits employed in the reproducing and monitoring 
operations are identical, it will be unnecessary to explain further the 
operation of the theater expansion equipment, a schematic diagram 
of which is shown in Fig. 9. The use of identical monitoring and 
theater reproduction circuits insures that the sound reproduced from 



10 



20 



30 



\ 



\ 



2000 3000 4000 

FREQUENCY 

FIG 8. Variable-mu stage gain vs. control-track frequency. 

the control- track print will have the same degree of expansion as 
that heard in the monitor during the re-recording process. Certain 
modifications are, of course, required in the sound-head for proper 
reproduction of the control- track. For example, the scanning aper- 
ture must be widened from the present 84-mil standard to the pro- 
posed width of 105 mils. This may be accomplished by widening 
the physical slit in the reproducing lens system or by increasing the 



Feb., 1942] CONTROL-TRACK FOR MOVIETONE PRINTS 



121 



magnification of the objective lens in the optical system. Fig. 10 
shows the modifications to a standard Western Electric TA-7400 
necessary for proper reproduction of the control- track. It will be 
noted that the light transmitted by the sound and control- tracks is 
separated into two distinct beams by two abutting lenses, the indi- 
vidual beams being transmitted to the active surfaces of two sepa- 
rate photocells, or the separate surfaces of a push-pull type of cell. 
If the latter technic is employed, special balancing PEC coupling 
circuits must be employed to reduce cross-talk between the signal 
and control-tracks. In certain types of sound-heads, where it might 
be difficult to mount an additional cell, the use of the push-pull PEC 
and associated balance circuits is probably indicated. The PEC 
output from the two tracks is fed over two separate coaxial cables to 
the special amplifying equipment previously described. The pre- 
amplifier preceding the variable-gain stage may not be necessary in 







VAR. 


GAIN 


"kj* 


AMPLIFIER 




EPRODUCE 

V * 




AMPLIFIER 












DISCRIMINA- 


v. -^ 




TOR 



FIG. 9. Theater expansion system. 

many theaters having modern single or two-stage pre-amplifiers 
operating from cable connections from the photocells located in the 
sound-heads. While the actual physical design and booth location 
of the special reproducer equipment can not be specified exactly at 
this time, the circuits employed will undoubtedly be quite similar to 
those that have been described above. 

In reproducing the sound-track in theaters which have not been 
equipped for reproduction of the control-track, some difficulty may 
be encountered from audible reproduction of the control frequencies 
over the theater horn system. This can usually be prevented by read- 
justing the position of the scanning slit. However, in the case of 
some of the older optical systems, it may be necessary to insert a 
mask at the scanning slit to prevent partial scanning of the control- 
track by the light-beam. In theaters employing modern optical 
systems, there should be no difficulty provided the lens system is in 
proper adjustment. 



122 



J. G. FRAYNE AND F. P. HERRNFELD [J. S. M. P. E. 



While the use of a single frequency-modulated control channel has 
been described here, consideration has also been given to the use of 
multiple channels superimposed on the same 5-mil track. It has 
been found that three channels each capable of approximately 30-db 
expansion and suitable for high-speed operation may be accommo- 
dated. The number of channels may be increased almost indefinitely 
if band width and time of operation are not limiting factors. These 
multiple channels may be used to control the outputs of multiple 
tracks intended for stereophonic or similar purposes or they may be 
used, if desired, to switch music or sound effects to multiple horns or 



STANDARD HEAD 



SLIT & SEPARATOR 



REFLECTOR 




FIG. 10. Modifications in WE TA-7400 for reproduction of the control-track. 

for other similar operations intended to add more realism to the 
screen. 

Recording Results. The use of this control-track has, to date, been 
limited to experimental re-recordings of sequences at Universal and 
Samuel Goldwyn Studios, and the re-recording of a 2-reel musical 
short at Universal, which is intended for demonstration at the Pan- 
tages Theater in Hollywood. In none of these tests was it found de- 
sirable to use less than about 12 db of expansion to secure the proper 
range of sound volume from the print. In others it was found de- 
sirable to use the full 30-db range. This was particularly true in the 
re-recording of a dive-bombing sequence in the picture, The Long 
Voyage Home, in which the top gain was required f or proper repro due- 



Feb., 1942] CONTROL-TRACK FOR MOVIETONE PRINTS 123 

tion of the bomb bursts and 30 db less gain was required for proper 
rendition of low-level background music passages. The fact that 
these extremely loud sounds can be recorded without overloading the 
light- valve makes possible their reproduction with a degree of natu- 
ralness and realism which heretofore could not be obtained on ac- 
count of the excess distortion incurred from overload of modulator 
and film. 

Conclusion. It has been found that a 5-mil frequency-modulated 
control-track recorded between the present sound-track and picture 
area on the release negative may be used to add 30 db of volume range 
to existing sound-films. It has also been found that with this device 
it is no longer necessary to use squeeze-track or printer-light control 
methods to extend the volume range of variable-density sound-films. 
The control-track may be recorded in such a manner that a part of 
the gain change, the upper 20 db, for example, may be employed to 
enhance the volume of the louder sounds, while the lower 10 db may 
be used to reduce the gain during the quieter passages ; thus adding 
effectively to the noise reduction during these intervals. The use of 
the control-track in the area specified is not limited to any particular 
operating speed nor is it limited to the operation of a single sound 
channel, but may be used also to provide controls for multiple sound- 
tracks. In addition it may be used for various other types of control 
operations associated with reproduction of sound-films. Standard 
sound-films may be reproduced in theaters equipped for control- 
track by simply switching in a fixed bias in the variable-gain stage 
amplifier, and control-track films may be played as standard tracks 
in unmodified theaters. Experience to date with the use of this track 
has shown that it may be used very effectively not only to enhance the 
realism of high-level sounds, but to add much to the dramatic quali- 
ties of low-level passages where the usual presence of background 
noise detracts from the scene being portrayed. 

REFERENCE 

1 LEVINSON, N., AND GOLDSMITH, L. T. : "Vitasound," J. Soc. Mot. Pict. Eng., 
XXXVH (Aug., 1941), p. 147. 

DISCUSSION 

MR. PALMER: Does the use of this proposed control-track involve the addi- 
tion of new amplifying equipment in the theaters? 

DR. FRAYNE: In addition to the modifications of the sound-heads outlined in 
the paper, the only change contemplated in theater equipment for proper repro- 



124 J. G. FRAYNE AND F. P. HERRNFELD 

duction of the control-track is the addition, if necessary, of sufficient amplifier 
capacity to provide proper reproduction of the louder passages. 

MR. KELLOGG: The criticism has been made of any system depending upon 
changing amplitude of a fixed frequency that it would be susceptible to unde- 
sired changes due to variations in exciter lamp brightness and photocell sen- 
sitivity (including effect of variations in polarizing voltage). I should like to 
call attention to the fact that this problem was early considered by H. I. Reiskind 
who handled most of the development work on the sprocket-hole control-track for 
us. Following a suggestion of Frank Sheppard, he worked out what has proved 
to be a very satisfactory way of overcoming this difficulty. Making use of the 
logarithmic relation between plate current and grid voltage when an amplifier tube 
is worked in a certain range, he caused the average photocell current to change 
the amplification so that the 96-cycle output of the tube depended upon the per- 
centage modulation of the transmitted light and was scarcely affected by a change 
in average brightness of the source, or photocell sensitivity. This work was re- 
ported in Mr. Reiskind's paper on "Multiple Speaker Systems" in the August, 
1941, JOURNAL. 

MR. FARNHAM : Is the picture area affected by the use of the control-track? 

DR. FRAYNE: The use of the control-track described in this paper does not 
call for any change in standard picture or sound-track areas. 



DESIGN AND USE OF NOISE-REDUCTION BIAS SYSTEMS* 
R. R. SCOVILLE AND W. L. BELL** 



Summary. The factors underlying the design and use of biased recording sys- 
tems are described. In order to minimize noise and "shutter bump" special precau- 
tions in filtering must be taken. Suitable values for "attack" and "release" times are 
dependent upon the type of recording, margin settings, and reproducing conditions. 
Comparison of variable-density and variable-area requirements is made. Methods 
used in designing the rectifiers, filters, and other circuit details are given and the appli- 
cation to a new equipment known as the RA-1124 Noise-Reduction Unit is shown. 

The reduction of film background noise by the use of biased re- 
cording modulators is widely used and the general principles are well 
known. 1 ' 2 However, many factors enter into the design and use of 
noise-reduction equipment which are not so well known. This 
paper will discuss the subject particularly with respect to variable- 
density film recording, but much of the information should also apply 
to variable-area systems. 

The improvement afforded by the so-called biased recording 
method depends upon the fact that noise is reduced when the amount 
of average light transmitted through the sound print is held to the 
lowest average value which will still accommodate the desired signal. 
In variable-density recording this effect is readily obtained by super- 
imposing a bias current on the regular signal current in a light-valve 
(or similar device) in such a manner as to partially close the opening 
during quieter intervals. For louder signals the bias current automat- 
ically changes in the direction of increasing the average valve opening 
to an extent which permits the modulator to vibrate without " clash" 
or overload. No volume distortion of the signal should be "caused 
thereby since the "signal component" of the emergent light is un- 
changed. Only the surplus light or "d-c component" is altered. 
Similarly in the variable-area system a bias is used to operate a 

* Presented at the 1941 Fall Meeting at New York, N. Y. ; received December 
1, 1941. 

** Electrical Research Products Division of Western Electric Co., Hollywood, 
Calif. 

125 



126 R. R. SCOVILLE AND W. L. BELL [J. s. M. p. E. 

shutter which masks off portions of the negative film not required to 
carry the signal amplitude. When printed such masked areas be- 
come black and thus reduce noise. Both methods are essentially the 
same having the common objective of reducing "surplus light." The 
principle involved is simple but difficulties arise because of two fac- 
tors: 

(1) The addition of undesired noise components by the working of the bias cur- 
rent. 

(2) The time delay between the arrival of a signal and the establishment of the 
corresponding bias current, which causes "clipping." 

(1.0) AUDIBLE EFFECTS PRODUCED BY NOISE REDUCTION 

(1.1) Low Frequency Noise. The extraneous signal produced by 
the bias current action must be limited by means of a filter to fre- 
quencies so low as to become inaudible over the reproducing loud 
speakers used. Sensitivity of the ear falls off at the lowest frequen- 
cies so that a suitable filter design becomes possible. In practice the 
design of such a filter becomes a compromise between the amount of 
noise which can be permitted to come through (see sec. 7.3) and the 
desired rate of build-up or "attack" time of the bias current. Obvi- 
ously the more rapid the "attack" the less will be the "clipping." If 
the attack is too rapid a "thump" or "shutter bump" will occur when 
a sustained signal is impressed upon the bias circuit. The use of 
push-pull records 3 reduces both "thumps" and other low-frequency 
noise brought in by the bias action. 

(1.2) Film Noise Modulation. Another form of noise more diffi- 
cult to deal with is noise modulation or "hush-hush." This is caused 
by the change in amplitude of film noise at a rate dependent on the 
bias current. Variable-density recording using film stocks prior to the 
advent of the new fine-grain types exhibited the effect to a marked 
degree. For example, piano music was most difficult to record satis- 
factorily because the envelope of this type of sound is constantly 
varying over a wide range at rates which pass through the noise-reduc- 
tion filter circuits. The reproduced film noise as a result rises and 
falls to produce a disturbing effect sometimes worse than a steady 
film noise of greater loudness. The term "hush-hush" has frequently 
been used in reference to the trailing noise which follows up and fills 
in the gaps in such recordings as piano music. This stands out be- 
cause it often is not masked by the desired sound. Along with this, 
and not so generally recognized as such, may be a more rapid noise 



Feb., 1942] NOISE-REDUCTION BlAS SYSTEMS 127 

modulation proceeding sometimes at rates up to ten or fifteen cycles 
per second. When recordings of single piano notes are made with 
biased recording systems there is some "clipping" during the first 
few milliseconds after each note is struck due to the time lag in the 
bias circuit. This does not seem to degrade the record quality ap- 
preciably, however. The recorded variable-density film may show 
some "hush-hush" of the type which trails along behind the note but 
this is generally not very serious. The most objectionable effects are 
obtained when chords or rapid sequences of notes are played. Ex- 
amination of oscillograms of this type of sound shows usually a rapidly 
varying envelope causing the bias current to fluctuate up and down 
about as rapidly as the filter will permit. The resulting fluctuations 
of film noise when the print is reproduced can be very annoying. 
Reducing the amount of the bias fluctuation by lengthening the re- 
lease time improves the record, but at the same time increases the 
average noise. Noise-modulation phenomena may occur indepen- 
dently of the bias effect. For example, if a low-frequency signal, such 
as 30 cycles, is recorded without noise reduction and is later repro- 
duced through a high-pass filter which excludes the fundamental 
signal an unpleasant sound is heard which is amplitude modulated 
film noise at the low-frequency rate. The conclusion is evident there- 
fore that noise modulation can be caused not only by the bias action 
but by the principal sound signals as well. Fortunately, the use of the 
new fine-grain sound-films 4 greatly reduces the magnitude of these 
noise-modulation effects so that recordings of the piano and other 
difficult subjects can be made today very satisfactorily by the vari- 
able-density system. 

Another method which is quite effective in minimizing noise- 
modulation consists in the use of "pre and post" equalizers. By this 
method the higher frequencies are preemphasized in recording and 
correspondingly attenuated in reproducing. The overall frequency 
characteristic is thus not changed but the film background noise is 
appreciably reduced. Variable-area records generally have less 
noise-modulation trouble than do variable-density tracks. One rea- 
son for this is that in variable-density the noise amplitude varies with 
the average transmission of the film, whereas in variable-area the 
noise amplitude varies with the square-root of the transmission. 
Another contributing factor is that the graininess of a variable-area 
track is less than that of a corresponding density track. These ad- 
vantages of the variable-area track are, however, to a considerable 



128 R. R. SCOVILLE AND W. L. BELL [J. S. M. P. E. 

degree offset by other types of noise caused by the presence of dirt, 
grease, scratches, pin holes, etc., which are more detrimental to area 
than to density tracks. 

(1.3) Volume Distortion Due to Film Characteristic. The forms 
of clipping described above are usually thought of as applying only 
to the modulating device. However, such clipping is considerably 
exaggerated at times by a non-linear film characteristic in variable- 
density recording. This occurs where the straight-line relationship 
between print transmission and negative exposure does not extend 
over a sufficiently long range. In the regions of low exposure to the 
negative the print transmission may become excessively dark and 
consequently the lower volumes are unduly reduced in amplitude. 
This may result in the record's having an explosive quality wherein 
the higher volumes are reproduced too loud in reference to the lower 
volumes. It may also give a feeling of deadness to the recording 
owing to the fact that reverberation and other low-volume effects 
are unduly attenuated. These effects are also caused by an overall 
gamma which is too high. An excessive setting of noise-reduction bias 
current also often gives this effect. One of the most satisfactory 
tests for the amount of noise reduction which is permissible is that 
furnished by the intermodulation test. 8 This is made by making 
such a test at a signal level 10 or 12 db below full modulation, with the 
noise-reduction system operating. Under such conditions the per- 
centage intermodulation should not greatly exceed that used for the 
normal full-modulation test. 

(1.4) Clipping. Ideally, noise-reduction bias currents would be 
applied by interposing a time delay in the speech circuit. Thus, there 
would always be sufficient carrying-capacity in the modulating device 
to accommodate the signal. Practically, this has not thus far proved 
feasible. A method has been developed called the "direct positive" 6 
which has the equivalent of anticipation but as this does not provide 
a negative and is not applicable in variable-density work, the method 
has not been very generally accepted. 

The type of clipping exemplified by single piano notes is to be ex- 
pected in the usual type of recording, but in general no serious dis- 
tortion results therefrom. 

A more serious form of clipping is incurred with sounds such as il- 
lustrated in Fig. 1. The upper trace represents the bias current and 
the lower shows the sound-wave, in this case speech. Here the time- 
lapse between successive peaks in the envelope is sufficient to permit 



Feb., 1942] 



NOISE-REDUCTION BIAS SYSTEMS 



129 



the bias current to release so that the oncoming succeeding peak is 
clipped and a whole progression of clipped peaks consequently results. 
The best way of avoiding this type of phenomenon is to make the re- 
lease time long enough to carry over from one peak to the succeeding 
peak. In practice this becomes a matter of making the release time 
as long as possible without unduly increasing the film surface noise 
owing to the greater average opening which is thereby present. Fig. 
2 shows a more suitable bias action in which a longer release time is 
used. The figures are sketches made from actual oscillograms, for 
better clarity of reproduction. 




FIG. 1. Bias action and clipping when release time is too short. (Lower 
trace) speech; (Upper trace) bias current. 




FIG. 2. Bias action with sufficiently long release time. 
( Upper trace) bias current 



(Lower trace) speech; 



(2.0) TIMING CHARACTERISTICS 

Attack time is arbitrarily defined as the time required for the bias 
current to undergo 90 per cent of its total change when a signal having 
a magnitude somewhat less than the value required to cancel the 
bias fully is applied to the noise-reduction unit. 

Release time is defined as the time required for the bias current to 
undergo 90 per cent of the change when the input signal applied as in 
the previous definition is removed. 

These definitions of the time-constants referred to in this discussion 
may not be comparable to the time-constants of filter sections de- 
scribed by others. Thus, in speaking of the time-constant of a 
capacity-resistance filter a value of C X R is frequently given as the 
time-constant. This represents the time required for the change to 



130 R. R. SCOVILLE AND W. L. BELL [J. S. M. P. E. 

proceed 0.63 or (1 1/e) of the total amount rather than the 0.90 
factor stated above. 

For variable-density recording of a single track using 10 db of noise 
reduction, an attack time of 20 milliseconds, as just defined, has been 
found generally satisfactory. Some studios, particularly those fa- 
voring a heavy low-frequency end, use up to 24 milliseconds while 
others work with as little as 15 milliseconds. 

Margin is defined as the difference in decibels between the applied 
signal amplitude and the amplitude which the signal could have if 100 
per cent modulation of the available spacing were used. In line-up 
work, margin is generally taken to be the difference in decibels be- 
tween the signal which just overloads the modulator and the signal 
which just cancels the bias current (or adjusts it to the final value). 

The attack time is the same for all values of input signal up to the 
value which just cancels the bias. If more than this input ampli- 
tude is impressed upon the noise-reduction unit the rate of change 
increases proportionally to the input signal but since the total change 
must remain the same a decrease in attack time results. Thus for 6 
db of over-cancellation (margin) the attack time is reduced to half. 

In order to minimize clipping the attack time is made as short as 
possible. The limit is determined by the "thump" obtained when a 
sustained signal is suddenly impressed. The loudness of the "thump" 
appears to be proportional to the rate of change of the average print 
density. On this basis the "thump" is loudest on maximum signal 
amplitudes and becomes less audible as the recording level decreases. 

In variable-area recording a somewhat different condition exists. 
For 10 db of noise reduction the space occupied by the clear area of 
the print must be reduced to 10 per cent of its normal width. This 
compares to variable-density recording in which light- valve spacing is 
reduced to about 32 per cent of normal. If it is assumed that the same 
rate of change of print transmission is tolerable for both systems, the 
attack time should be longer for area in the ratio of 90 to 68 per cent, 
or 1.32. Owing to the reduced initial spacing employed in variable- 
area recording a greater margin is needed than for variable-density. 
Thus, 4 to 8-db margin is often employed in area systems. Assuming 
for computation that only 2 db more margin is needed for area than 
for density, a correction factor of 1.25 would express the ratio of tol- 
erable attack time for area compared to density on the basis of margin 
only. Combining this correction with that due to the difference in 
amplitude of change described above would result in a total time of 



Feb., 1942] NOISE-REDUCTION BIAS SYSTEMS 131 

20 X 1.32 X 1.25, or 33 milliseconds for variable-area. This repre- 
sents the attack time for input signals which do not quite cancel the 
bias. For the maximum input signal, the time drops to about 20 
milliseconds. 

In a well balanced push-pull recording system the "shutter bump" 
and other noise components in the bias system are balanced out so 
that a faster attack time becomes feasible. Since clipping is re- 
duced by faster attack, there is an advantage in so adjusting. How- 
ever, some users have preferred to keep the timing the same for push- 
pull as for standard so that if only one of the two tracks is played, 
satisfactory reproduction will be obtained. In this case then, the 
only advantage which push-pull gives over standard recording is in 
the reduction of harmonics and some bias noise. A push-pull attack 
time of 8 to 10 milliseconds has proved very satisfactory for most 
push-pull recording work. The release time for push-pull density re- 
cording is generally the same as that employed with single-track re- 
cording. 

The choice of release time is a compromise, between the greatest 
reduction of noise on the one hand and freedom from bias action and 
clipping on the other. The longer the release time is made the less 
danger there is of clipping, since the spacing of the modulating device 
is held open longer, awaiting the arrival of oncoming wave-trains. 
Also, a long release time insures a reduction of the working of the bias 
current thus further reducing miscellaneous bias noise. However, 
the long release times cause more noise to be present during the in- 
terim periods than would otherwise occur. In variable-density sys- 
tems long release times have not been favored on this account. An- 
other factor which enters is reverberation of the room in which the 
recording is made. Where considerable reverberation is present, there 
is no necessity that the bias current die down quickly since the sound 
in the room does not die down quickly. Theoretically, therefore, the 
release time would vary with the reverberation time but since this is a 
somewhat impracticable adjustment to be made a compromise is taken 
which seems to give the best operation for most types of recordings. 
In variable-density a release time of 40 to 50 milliseconds appears to 
be a good compromise. In variable-area recording a longer value of 
about 150 milliseconds has generally been preferred because the dan- 
ger of clipping seems more serious with shorter times than is the case 
with density. 

The use of fine-grain films for variable-density sound may make 



132 R. R. SCOVILLE AND W. L. BELL [J. S. M. P. E. 

release times longer than 50 milliseconds desirable but inadequate 
data are thus far available on this point. The long release time has 
the advantage that simpler types of filters can be used than for short 
release systems, reliance being made upon the long release time to 
provide the required degree of filtering. 

(3.0) SHAPE OF ATTACK AND RELEASE CURVES 

Where simple capacity and resistance filters are used, the shapes 
of the attack and release curves obtained are exponential. Where 
full-section inductance and capacity filters are used slightly under- 
damped, the attack and release curves are more nearly linear, as in A 




VOLTAGE IMPRESSED ON FILTER 




FIG. 3. Characteristic attack and release curves of constant-J type filter 
when a current pulse is impressed. 01) For under-damped condition but no 
overshoot. (B} For critical damping. (C) For serious under-damping with 
excessive overshoot. 

of Fig. 3. It is of interest here that such filters if critically damped 
are exponential as in the first case. When not critically damped, 
however, an additional component is added having an oscillatory char- 
acteristic which has the effect of more quickly building up the attack 
current to its full value, generally with some slight overshoot. This is 
illustrated in curve C of Fig. 3 for a badly under-damped case. The 
question has frequently been brought up as to whether the shape of 
the attack curve should be exponential or linear. The answer is that 
the shape has a characteristic relationship to the filtering efficiency 
and the one can not be changed without affecting the other. Thus, for 
any given "attack" and "release" time the most effective filtering is 
obtained with full-section K or M-type filters slightly under-damped 
and these are inherently more nearly linear than exponential. This is 
further discussed in the section on filtering. In variable-area work 



Feb., 1942] 



NOISE-REDUCTION BIAS SYSTEMS 



133 



the exponential shape has generally been used with satisfactory re- 
sults, whereas in variable-density the linear shape has been more 
common, largely because of the characteristics of the filters used. 

The shapes of both the attack and release curves are altered by the 
amount of margin (over-cancellation) used. Thus, the attack time is 
reduced by an increase in margin as previously discussed and the re- 
lease time is increased. The release characteristic exhibits the valu- 
able feature of remaining fully open for an appreciable period after 
the cessation of the over-cancelling signal, after which it falls off nor- 
mally. The preferred shape of the release characteristic would seem 
to be exponential. This is because the noise should fall off uniformly 
as the release period progresses. A small change in modulator cur- 
rent near the end of the release period makes a relatively large change 
in noise and therefore the rate of change of current should be rapid 



01 SEC 



ATTACK TIME 



RELEASE TIME 




FIG. 4. Typical oscillogram of attack and release currents considered suitable 
for variable-density single-track records. 



at the beginning of the release period and progressively smaller as it 
reaches the end. This effect is obtained by the use of a capacity-re- 
sistance type of filter or the combination of a capacity-resistance 
filter and an M or K-type section, as will be further described later. 
Fig. 4 shows an oscillogram of the way in which current builds up 
and releases with such a circuit suitably adjusted. This is considered 
to represent the most desirable form, at least for variable-density re- 
cording. 

The transient effect present due to under-damping of the circuits 
(illustrated in Fig. 3) is arbitrarily held to less than a 5 per cent over- 
shoot. It is doubtful whether there is any particular disadvantage 
in a moderate overshoot. 

(4.0) NOISE REDUCTION AND MARGIN 

The term Decibels Noise Reduction has been quite widely used, par- 
ticularly in the variable-density system. It indicates the expected re- 



134 R. R. SCOVILLE AND W. L. BELL [J. s. M. P. E. 

duction of background noise by the use of bias currents. In variable- 
density systems the noise output varies approximately linearly with 
the mean projected transmission of the sound-track, thus: 

7* 

Db Noise Reduction = 20 log 



where T is equal to the mean transmission of the print in the un- 
biased condition and T b is equal to the transmission of the film when 
the bias is applied. Since the transmission of the print is proportional 
to the exposure of the negative, one may substitute for the trans- 
mission the corresponding spacings of the modulating device with and 
without bias current, respectively. 

The variable-area noise output varies in proportion to the square- 
root of the clear track width so that here : 

r> 

Db Noise Reduction = 10 log -^ 

Sb 

where 5 is the width of the clear area of the print in the unbiased 
condition and 5& is the width of the clear area when the bias is applied. 

These expressions indicate, for example, that with 10 decibels of 
noise reduction, 10 per cent of full opening in the variable-area sys- 
tem would be used, whereas in the variable-density system 31.6 per 
cent would be used. In practice, the bias for variable-area systems is 
generally set in reference to the width of the bias line rather than by 
referring to decibels of noise-reduction. 

The amount of noise-reduction and margin used at the various 
studios employing the variable-density recording method varies over 
a considerable range. This is shown in Table I. 

TABLE I 
Average Settings of Noise-Reduction Equipments Variable- Density Method 

Noise Reduction Used 
Type Record Margin Range Average 

Single-Track Original Dialog 2-3 db* 10-13 db 12 db 

Single-Track Original Music 4 db 8-9 db 8 db 

Push-Pull Original Dialog Va-3 db 9-13 db 14 db 

Push-Pull Original Music V 2 -3 db 10-18 db 14 db 

Single-Track Release 2-6 db 10-13 db 12 db 

* These values apply to studios using peak type noise-reduction units. The 
older average response type units employ 4 to 6 db margin. 



Feb., 1942] NOISE-REDUCTION BIAS SYSTEMS 135 

The values shown represent total noise-reduction and include that 
portion obtained by the reverse bias method. The latter method is 
explained as follows: A nominal amount of noise-reduction is com- 
puted on the basis of closing the valve down from the zero bias condi- 
tion, as, for instance, 8 to 15 db. A further bias in the reverse direction 
is provided so that on signal amplitudes larger than average the bias 
current is made to open the light- valve from 3 to 6 db (50 to 100 per 
cent) further. This opening increase is generally permitted to intrude 
in exposure upon the non-linear portion of the film characteristic and 
is sparingly used in recording. Since the overload characteristic of 
film is gradual the harmonic distortion incurred by so doing is not 
great and an appreciable increase in "ceiling" is thus obtained. 

Many users of the Western Electric variable-density system prefer 
the push-pull method for all original recording. On dialog only about 
2 db more noise-reduction is used here than for standard recording, 
whereas on music about 6 db more noise-reduction is obtained. The 
reduced distortion is considered of greater importance than the in- 
creased noise-reduction obtainable. The margin used varies from 
about 1 db to as much as 6 db with an average of approximately 3 db. 

(5.0) PEAK RESPONSE CHARACTERISTICS 

Some of the older noise-reduction equipment used for variable-den- 
sity work had a bias current output characteristic proportional to the 
average value of the input signal. It was found feasible in later cir- 
cuits, though at some cost in simplicity, to make this response pro- 
portional to the peak value of the input signals. This characteristic 
proved superior to the former one, because many sounds have highly 
peaked wave-shapes. Whereas a margin value of about 6 db had been 
necessary with the older-type circuits, a margin of 1 or 2 db appeared 
to be equivalent with the new. In order to obtain this peak response 
feature a relatively low-impedance circuit drives a rectifier, as in Fig.. 
9. This charges up a condenser (Co) in a time of 1 millisecond or less. 
Since this condenser must remain charged for an appreciable length 
of time wherein the bias current output may build up to the required 
value, a further specification calls for a relatively long release time of 
discharge of the said condenser. Therefore, the resistance R is made 
large compared to R Q in the figure. A ratio of approximately 100 to 1 
in the discharging and charging impedances of this storage condenser 
seems desirable. Since then the ratio of the terminating resistance and 
the generator resistance is large a very poor power transfer efficiency 



136 R. R. SCOVILLE AND W. L. BELL tf. S. M. P. E. 

is obtained in such circuits. In other words, only about 4 per cent of 
the energy fed to the rectifier can be utilized in final output. This 
need not necessarily be a serious obstacle to the use of such circuits 
since the rectifying may be done in relatively high-impedance low- 
level grid circuits, the voltage output of which is subsequently ampli- 
fied. 

(6.0) FULL-WAVE VS. HALF-WAVE RECTIFICATION 

Experience in variable-density recording has seemed to favor the 
use of full-wave rectification though some users of variable-area have 
preferred the half -wave scheme. 2 It is granted there is no necessity of 
bias action except on peaks which tend to close the modulating aper- 
ture, and some have argued that the bias should change in response 
only to such peaks as tend to close the aperture. The point is also 
made that most speech waves are dissymmetrical with the highest 
amplitude peaks nearly always on one side of the wave. Conse- 
quently, such signals may be polarized, it is said, so that the largest 
amplitude always produces an opening of the modulator, a direction 
in which bias current is of no importance. The peaks on the other 
side of the wave are then permitted to operate the half -wave rectifier 
to produce the necessary change in bias current for the small ampli- 
tude half of the wave. Although this method apparently has been 
made to work satisfactorily in some recording systems, there are sev- 
eral disadvantages which have thus far prevented recommendation 
of the scheme in Western Electric Sound Systems. Some of these 
objections are as follows: 

(1) Accurate poling of the microphone is required. This poling, although often 
reliable, is sometimes reversed by the nature of the pick-up material and by rever- 
beration in the room. Also, it is somewhat difficult to determine with some record- 
ing systems. 

(2) Low frequencies are less adequately filtered in the noise-reduction circuit. 
A full-wave rectifier doubles these frequencies and renders them more readily re- 
moved by the filter in the circuit. Curves A and B in Fig. 5 show the comparative 
transmission of noise frequencies through the filter for full-wave and half-wave 
rectifiers, respectively. In variable-density recording where both low attack 
and low release times are required in order to avoid hush-hush the filtering is often 
a very critical factor. 

(5) Use of the full-wave rectifier may often give an effective head start to the 
bias build-up since it starts the bias acting in advance of actual need. Thus, some 
reduction in clipping may be obtained over the half-wave system. The idea of 
poling the light-modulator so that the highest amplitude peaks open the device is 
undoubtedly to be recommended even where full-wave rectifiers are used. By so 



Feb., 1942] 



NOISE-REDUCTION BIAS SYSTEMS 



137 



doing a greater margin of freedom from clipping is obtained, but if the dissym- 
metry reverses for any reason there will be no damage done. Few if any users of 
light-valve systems employ this poling technic at the present time, probably be- 
cause the slight improvement gained is considered not worth the trouble. 

(7.0) CIRCUIT CONSIDERATIONS 

(7.1) General Design. The design of the circuit to be used for 
noise-reduction bias systems depends upon the impedance of the 
modulating device and the current requirements. Where the modu- 
lator has a medium or high impedance with a resulting low-current 



-5 

-10 
H5 
-20 
-25 

g-30 
-35 
-40 
-45 
-5O 




































UNFILTEREO COMPONENTS REF. TO SIGNAL 












































\ 


































\ 




























\ 






\ 


























\ 






\ 


B 


























\ 






\ 




























V 


\ 






























\ 


\ 
































\ 


\ 
































5 

















































10 20 100 

FREQUENCY SIGNAL IN CYCLES PER SECOND 



FIG. 5. Comparative filtering effectiveness of using (.4) full- 
wave and (B) half -wave rectifiers with associated network for noise- 
reduction bias. 



requirement, it may be operated directly in the plate-circuit of a 
vacuum-tube or out of a copper-oxide rectifier. 

Where very low-impedance modulators are used it is not practicable 
to use plate-current directly for bias owing to the large current re- 
quired. The use of copper-oxide rectifiers also presents difficulties for 
the following reasons: 

(I) The filters must be in low-impedance circuits so that the values of capacity 
required become hundreds and even thousands of microfarads which are difficult 
to obtain. Electrolytic condensers have been used but have disadvantages. 

(2} The peak response feature which necessitates a short attack and a long re- 
lease time is difficult to obtain effectively in copper-oxide circuits. 



138 



R. R. SCOVILLE AND W. L. BELL 



[J. S. M. P. E. 



The best solution for the low-impedance case seems to be the use 
of a modulated high-frequency carrier signal, the output of which is 
caused to vary in accordance with the peak value of the signal wave. 
A circuit of this type is described in Sec. 8.0. 

The method of connecting the bias output to the modulating de- 
vice becomes important where signal and bias are simultaneously im- 
pressed on a common modulator. One method is shown in Fig. 6 in 
which a blocking condenser is used to prevent the direct current from 



SOUND 
SOURCE 




i 


j " l| < 

1 * 

3 


! 
ij 




I 






f i 


i 
i 
1 

1 


NOISE 
REDUCTION 
UNIT 









LIGHT 
MODULATOR 



UNO 
RCE 







r 





i i 







c 






1 




NOISE 
REDUCTION 
UNIT 


* 







LIGHT 
MODULATOR 



FIG. 6. (Upper) Early method of combining bias and signal currents. 
FIG. 7. (Lnwer) Simplex method of combining bias and signal currents. 



passing through the transformer winding, 
jections : 



This method has two ob- 



(1) The condenser must have several thousand microfarads capacity where 
low-impedance modulators such as ribbon light-valves are used. 

(2} The impedance of the bias circuit must be high to avoid shunting the signal 
circuit by the rectifier in the bias system. 

A better method is that shown in Fig. 7 and generally known as a 
simplex circuit. The circuit is balanced by use of the dummy re- 
sistor R d , which is made equal to the resistance of the light- valve. In 
this way the bias and signal circuits can have no undesirable reac- 



Feb., 1942] 



NOISE-REDUCTION BIAS SYSTEMS 



139 



tions upon each other. The simplex method requires twice as much 
current from the bias circuit as the valve alone requires and the net 
impedance is half. Thus, for the usual single-track light-valve sys- 
tems the impedance is only about one-half ohm and the current 
requirement is in the neighborhood of 500 ma, where both direct and 
reverse bias are used. 

In push-pull systems the dummy resistor may be replaced by the 
second light-valve. 

(7.2) Design of Filter Circuits. The simplest type of filter consists 
of a capacity on the output of a rectifier discharging into a resistance. 
This type of circuit, which has been analyzed by Kellogg, 6 can be 
made to give any attack or release time and has an exponential build- 
up and release characteristic. Its chief disadvantage is its relative in- 



' FILTER SECTION 





R 




L R 






-* 


rWVW 





onmnnnrp AAAAA 





. 


TO 


1 




, 




^> r\\ IT 


RECTIFIER 


_L 








^ o OU 1 

< T 




T C 




~ C Z 


~ C 3 


T 




1 










^ 


1 











OUTPUT VOLTAGE 
TO MODULATOR 



FIG. 8. Constant-^ filter with suitable capacity-resistance input network for 
noise-reduction filtering. 

efficiency in filtering out unwanted noise components in the bias cur- 
rent as compared with inductive-capacity types of filters. This fea- 
ture may not be a sufficiently important reason to discredit the simple 
capacity-resistance filter where it is possible to use relatively long re- 
lease times. 

The constant-X type of filter section, 7 as in Fig. 8, gives excellent 
filtering efficiency and has been extensively used in variable-density 
systems. When terminated with equal input and output impedances 
the attack and release times are equal, an undesirable condition as 
pointed out previously. However, this can readily be overcome by 
adding a section of capacity-resistance filtering as shown in Fig. 8. 
Here the input and output impedances are the designed values for the 
.K-type filter-section but the timing is modified by the input capacity 
Co in a useful way. If the generator impedance RQ is small compared 
to the circuit impedance R and also if R Q C Q is small compared to the 
attack time of the .K-type filter-section, the effect of the condenser 



140 



R. R. SCOVILLE AND W. L. BELL 



[J. S. M. P. E. 



CQ may be disregarded in its effect on the total attack time. It will, 
however, have a marked effect on the release time since the discharge 
of this condenser must be through the two resistors of value R and the 
filter elements. Where the product of CQ times 2R is large compared to 
the release time of the K-type filter the value of condenser CQ becomes 
almost the sole determining factor for the release time and the shape 
of the curve is exponential, as is desirable. This circuit also lends it- 
self to peak response action since the condenser CQ may be made to 
charge up very rapidly and to hold a charge while the final output 
voltage is building up in the remainder of the circuit. 

If a standard design of j?-type filter, as illustrated, is used in which 
Cz equals 3, an oscillatory condition will generally be obtained. 
In order to damp this circuit satisfactorily, resistance may be added 



DRIVER STAGE 




OUTPUT VOLTAGE 
TO MODULATOR 



FIG. 9. 



Modified M -derived filter with capacity-resistance input network as 
used in RA-1124 type noise-reduction unit. 



in series or across the inductance or in series or across each condenser. 
However, a more desirable method consists in distributing the ca- 
pacity unequally between Cz and C 3 so that the ratio of the capacity 
between the two is 3 or more as required for damping. It makes no 
difference here which of the two condensers is the larger. 

The X-type filter has a disadvantage in that the attack period is 
generally rather slow in commencing after the arrival of a signal. To 
overcome this a condenser by-pass across the inductance may be ap- 
plied with a suitable resistance in series to damp out the transient 
otherwise introduced at this point (see Fig. 9). This was found to 
improve the starting characteristic greatly, as shown in Fig. 10, and 
also to give a more desirable filtering characteristic because the 
resonance frequency of the parallel circuit could be adjusted to that 
value where additional filtering is chiefly needed. This is generally 
at about 50 cycles. It is apparent that this type of structure is very 



Feb., 1942] 



NOISE-REDUCTION BIAS SYSTEMS 



141 



similar to an Jkf-derived type of low-pass filter 7 in which the frequency 
of "infinite" attenuation is about 50 cycles and the cut-off frequency 
about 25 to 30 cycles. Thus the following expressions give the design 
values for the elements : 

mR 



m 



= Vl-/.V/ 



where R 
Ci 
f. 

/CO 

Cz + C 2 



circuit impedance 

capacity across series arm 

cut-off frequency 

frequency of "infinite" attenuation 

total shunt capacity of the section 



Here Cz and C 3 may be divided so that approximately one-third of the 
total is placed at one end of the section and the other two-thirds at 
the opposite end as required for suitable damping. It should be em- 
phasized that this circuit is not critically damped but is under-damped 




SLIGHT OVERSHOOT 



RELEASE 



VOLTAGE IMPRESSED ON FILTER 



FIG. 10. Showing how use of capacity (.4) across series inductive arm of 
filter improves attack curve as compared with X-type filter (5). 

to an extent necessary to give an approximate linear attack character- 
istic. The value of resistance added in series with condenser C\ is such 
as to limit the transient effect in the L\C\ circuit as can be determined 
from oscillograms of the action. Generally, this resistance will be 
about 10 per cent of the circuit impedance. The attack time in this 



142 



R. R. SCOVILLE AND W. L. BELL 



[J. S. M. P. E. 



type of circuit is proportional to v Li(Cz + C 3 ) and is very approxi- 
mately equal to l /zf c - 

(7.3) Measurement of Filtering. The term "Filtering" as used in 
reference to noise-reduction circuits described herein is denned as the 
difference in decibels between the amplitude of the fundamental sig- 
nal in the modulating device and the noise amplitude of the bias com- 
ponent produced by the impressed signal. Since such a difference 
varies with the input and bias current the definition is said to apply to 
10 db noise reduction and 6 db margin, and the filtering is measured at 





































e 

I 

;i 

1- 

o *** 

8 
2 

E 

LJ 

b 
i 








\ 


\ 






























\ 


\ 






























\ 


\ 
































\ 


K A 




























V 
































\ 


\ 


^v 






























\ 


s 


^ 






























s. __ 


, \ 
































\\ 


































\ 


































^ 


- 













10 20 50 100 200 

SIGNAL FREQUENCY IN CYCLES PER SECOND 



FIG .11. Relation between unfiltered components in bias current 
and recorded signal as per definition of filtering. Condition of 10- 
db noise reduction, 6-db margin, half -cancellation of bias; (A) for 
push-pull timing (0.01 sec. attack, 0.04 sec. release) ; (B) for single- 
track timing (0.02 sec. attack, 0.04 sec. release). 



that input condition where half-cancellation occurs. Starting at a 
frequency of 20 cycles, the output to the modulating device is first 
measured with both signal and bias currents present. Then the input 
signal is disconnected from the modulator and the noise components 
are measured. The difference between the two values (assuming a flat 
frequency characteristic) can then be computed. Other methods of 
measuring filtering are used which may be more convenient but the 
result should be the same. Fig. 1 1 shows, typical filtering character- 
istics for (A) push-pull and (B) standard recording systems, respec- 
tively. 



Feb., 1942] NOISE-REDUCTION BlAS SYSTEMS 143 

(8.0) A NEW NOISE-REDUCTION BIAS UNIT 

An all-purpose noise-reduction bias unit, making use of the above- 
mentioned considerations, has recently been built. This will deliver 
sufficient bias to give closure on any Western Electric light- valve cir- 
cuit and will also operate the Western Electric variable-area shutter. 
The timing is easily changed for standard or push pull track for either 
variable-area or variable-density records. Peak-type operation is em- 



20KC OSCILLATOR 




FIG. 12. Simplified schematic of RA-1124 noise-reduction unit. 

ployed. The output current is very stable in the presence of moderate 
supply-voltage variations. The operation is based on the carrier-type 
circuit, several examples of which have been described in previous 
literature. 9 - 10 

A simplified block diagram of the circuit used is shown in Fig. 12. 
It consists of an audio signal amplifier with a variable-gain control 
and equalizer, followed by a full-wave rectifier and filter which gives 
the desired timing and frequency characteristics to the rectified audio 
signal. This signal modulates the twenty kilocycles generated by the 
oscillator. The 20-kc amplifier amplifies the modulated signal which 



144 



R. R. SCOVILLE AND W. L. BELL 



tj. a M. P. E. 



is rectified, filtered, and supplied to the output as the noise-reduction 
bias current. 

The input circuit is unbalanced, and has an impedance of about 
5000 ohms so as to give less than one-half decibel bridging loss across 
a 500-ohm bus. An equalizer having a frequency characteristic simi- 
lar to that of the light- valve can be connected in the circuit when used 
with light- valve recording systems to give the same frequency char- 
acteristic to the signal operating the noise-reduction unit as is effec- 




FIG. 13. RA-1124 noise-reduction unit; front view. 



tive in the modulation of the light-valve, or disconnected for modu- 
lators having a flat characteristic. 

The signal amplifier uses a WE 348-A vacuum-tube for the gain 
stage followed by a WE 349-A power pentode for power. A 0.5- 
decibel-per-step potentiometer controls the gain of the audio signal 
amplifier, and is used in operation to set margin. Feedback is used 
over the two stages so that the gain is stable, and the output im- 
pedance is low. This amplifier overloads sharply and acts as a peak 
chopper at a few decibels above full cancellation of the bias output. 
This minimizes changes in operating times with audio signals greater 
than those necessary to cancel the bias to zero. The signal am- 
plifier output is rectified by a WE 3 51- A full- wave diode power rec- 



Feb., 1942] 



NOISE-REDUCTION BIAS SYSTEMS 



145 



tifier, with a relatively low internal impedance. The low output 
impedance of the amplifier and the high impedance of the timing 
filter circuit contributes a close approach to peak operation. 

A timing filter with an R-C section followed by an M-derived sec- 
tion with enough dissipation to eliminate undesirable transient effects 
is used. For normal light-valve recording an attack time of 20 milli- 
seconds and a release time of 40-50 milliseconds are used. For push- 
pull variable-density an 8-10 millisecond attack and a 40-50 millisec- 
ond release are employed. These filters are mounted on "plug-in" 
bases, so filters with different characteristics can be used as desired. 




FIG. 14. RA-1124 noise-reduction unit; rear view. 

The output of the entire unit must be stable with supply-voltage 
variations. To accomplish this the output of the oscillator is regu- 
lated. This is achieved by using a stabilized R-C oscillator, the gen- 
eral principles of which are well known. The circuit actually used 
was one developed by the. Bell Telephone Laboratories and has the 
advantage of simplicity and good stability. The Wien bridge deter- 
mines the frequency of oscillation and the thermo-resistor stabilizes 
the output level. 11 

The modulator uses a WE 348- A vacuum-tube with negative grid- 
bias modulation, and the output is taken across the cathode resistor. 
The modulator output is stabilized relative to supply-voltage changes 
by obtaining grid bias from a voltage-divider across the high- voltage 
supply, and adjusting it to reduce the plate current just to zero with 



146 R. R. SCOVILLE AND W. L. BELL [j. S. M. P. E. 

no signals on the grid. At this adjustment the voltage on the plate 
is proportional to t*e g where ju is the amplification factor and e g is the 
grid voltage. Any fluctuation in supply voltage shifts the grid bias 
an amount which just compensates for the changed plate voltage and 
the signal output remains substantially constant. 

A threshold adjustment is given by returning the rectifier cathode 
to ground through an adjustable voltage-divider. This puts a buck- 
ing voltage in series with the timing-filter output so that audio signals 
below a desired level do not begin to cancel the bias. This enables 
the unit to be set up to operate with approximately constant margin 
as the bias varies in operation from maximum to minimum. Where 
"hush-hush" troubles are obtained this feature is sometimes used to 
advantage. However, for most work the threshold device is not used 
since it increases clipping. 

The modulated 20-kc signal is amplified by a two-stage amplifier 
consisting of a WE 348-A gain stage, and a WE 349-A vacuum-tube 
power-output stage. A potentiometer controls the amplifier gain, and 
in operation is used to set the bias current. The output transformer 
has a one-ohm output winding for light- valve operation and a ten- 
ohm winding for area modulators. The desired winding can be con- 
nected to a full-wave copper-oxide rectifier and carrier filter, so that 
the envelope signal is supplied to the output as bias current. This 
unit will supply in excess of 0.8 ampere of bias current to a one-ohm 
load, or 0.25 ampere to a ten-ohm load. Provision is made to con- 
nect a d-c supply to the output circuit to furnish a reverse bias to the 
light-modulator. Photographs of the unit, which has been coded 
RA-1124, are shown in Figs. 13 and 14. 

CONCLUSION 

. This paper has attempted to cover the salient points in the design 
and use of biased recording systems. The subject is a complex one 
and the authors are indebted to many individuals and especially to 
Mr. John Livadary for many helpful suggestions, original ideas, and in- 
formation along this line. 

REFERENCES 

1 FRAYNE, J. G., AND SILENT, H. C. : "Western Electric Noiseless Recording," 
J. Soc. Mot. Pict. Eng., XVIII (May, 1932), p. 551. 

2 KELLOGG, E. W.: "Ground-Noise Reduction Systems," /. Soc. Mot. Pict. 
Eng.. XXXYI (Feb., 1941), p. 137. 



Feb., 1942] NOISE-REDUCTION BlAS SYSTEMS 147 

3 FRAYNE, J. G., AND SILENT, H. C.: "Push-Pull Recording with the Light- 
Valve,'* /. Soc. Mot. Pict. Eng., XXXI (July, 1938), p. 46. 

4 DAILY, C. R., AND CHAMBERS. I. M.: "Production and Release Applica- 
tions of Fine-Grain Films for Variable-Density Sound Recording," /. Soc. Mot. 
Pict. Eng., XXXVIII (Jan., 1942), p. 45. 

5 DIMMICK, G. L., AND BLANEY, A. C.: "A Direct Positive System of Sound 
Recording," /. Soc. Mot. Pict. Eng., XXXIII (Nov., 1939), p. 479. 

6 KELLOGG, E. W., AND PHELPS, W. D. : "Time-Delay in Resistance- Capacity 
Circuits," Electronics (Feb., 1937), p. 22. 

7 SHEA, T. E.: "Transmission Networks and Wave-Filters," D. Van Nostrand 
Co. (1929), pp. 221-291. 

8 SCOVILLE, R. R., AND FRAYNE J. G. : "Analysis and Measurement of Dis- 
tortion in Variable- Density Recording," J. Soc. Mot. Pict. Eng., XXXII (June, 
1939), pp. 648-672. 

9 HOPPER, F. L., MANDERFELD, E. C. f AND SCOVILLE, R. R.: "A New High- 
Quality Film-Recording System," /. Soc. Mot. Pict. Eng., XXVIII (Feb., 1937), p. 
191. 

10 HOPPER, F. L., MANDERFELD, E. C., AND SCOVILLE, R. R. : "A Light- Weight 
Sound-Recording System," /. Soc. Mot. Pict. Eng., XXXIII (Oct., 1939), p. 449. 

"PEARSON, G. L.: "Thermistors, Their Characteristics and Uses," Bell 
Laboratories Record, XIX (Dec., 1940), p. 106. 



A PRECISION DIRECT-READING DENSITOMETER* 
MONROE H. SWEET** 

Summary. A photoelectric densitometer of the direct-reading type is described. 
A logarithmic amplifier circuit has been modified so as to give an accurately linear 
output with high stability. The instrument covers a density range of to 3.0. Per- 
tinent optical, electrical, and performance factors are discussed. 

INTRODUCTION 

Historical Background of Objective Densitometer s. Ever since 
Hurter and Driffield's classic investigations of the characteristics of 
photographic emulsions, there has been a genuine need for accurate 
instruments to measure photographic densities. The state of develop- 
ment of optics and electricity during Hurter and Driffield's time neces- 
sitated visual instruments, and until relatively recently these have 
been developed almost to the complete exclusion of instruments 
using physical detectors. However, the advent of barrier-layer 
photocells and the development of vacuum phototubes paved the way 
for practical objective instruments. All densitometers of this class 
have several inherent advantages over subjective meters. 

Present State of Densitometry. When the present situation is sur- 
veyed, one finds that even now, visual densitometers overwhelmingly 
outnumber objective instruments. The chief reason for this seems to 
be that there is no commercially available photoelectric densitometer 
which is sufficiently simple, inexpensive, and satisfactory for general 
laboratory use. On the other hand, a number of photoelectric den- 
sitometers are described in the literature, but most of them suffer from 
one or more of the following objections : 

(1) Too complex: Usually, the electrical circuit must be made 
complex in order to achieve high gain with reasonable stability. This 
necessitates several balance, calibration, and zero adjustments. 
Furthermore, such circuits are always a potential source of difficulty 
and require expert servicing in the event of circuit trouble. 

* Presented at the 1941 Fall Meeting at New York; received October 20, 1941. 
** Agfa Ansco Research Laboratory, Binghamton, N. Y. 

148 



DIRECT-READING DENSITOMETER 



149 



(2) Too elaborate: In an effort to fulfill preconceived (and some- 
times unnecessary) requirements, some instruments have been 
made more elaborate than necessary. Often, the expedients used 
have seriously limited the light available to the phototube. This 



a. 

SPECULAR. 
DENSITY 



PHOTO CELL. 




PHOTOCELL AFFECTEP ONLY BY SPECULAR EKO.&ENT 



Jb 

DIFFUSE 




PHOTOCELL RESPONDS TO ALL EMERGENT LIGMT 




DIFFUSE 
INCIDENT LIGHT 

C 

ALTERNATIVE 
ARRANGEMENT 
FOR MEASURING 
DIFFUSE DENSITY 

PHOTOCELL AFFECTED ONLY BY SPECULAR. EMERGENT LIGMT. 

FIG. 1. Fundamental optical systems for measurement of density. 

creates the need for amplifiers of extreme sensitivity with their at- 
tendant difficulties. 

(3) Insufficiently accurate: Because of inherent instability, lack of 
suitable voltage regulation, or improper optical and electrical design, 
many densitometers are not capable of high accuracy. 



150 M. H. SWEET [J. S. M. P. E. 

(4) Cramped scales: Direct-reading instruments require a linear 
density scale to avoid excessive crowding at the high densities. Re- 
sort to a multi-range output indicator involves the inconvenience and 
confusion of range changing. 

(5) Too expensive: The technical demands placed on an acceptable 
general purpose instrument are quite severe. For this reason, most 
densitometers have been either compromised in design or are pro- 
hibitively expensive, i. e., they are not in the visual densitometer 
price class. 

Other less important objections become apparent as the instruments 
are used under working conditions. 

Measurement of Density. Optical density was defined originally by 
Hurter and Driffield as the log 1/T, where T is the transmission. Un- 
fortunately, no specification as to the quality or direction of the in- 
cident light was mentioned nor was the sensitivity and positioning of 
the receiver. A film specimen that is illuminated by collimated light 
and examined specularly is said to be measured as to specular den- 
sity. Such an arrangement is illustrated in Fig. l(a). When the 
receiver is located so as to intercept all of the emergent beam (see Fig. 
1&), diffuse density is measured. An alternative arrangement for 
measuring diffuse density is to illuminate the sample by diffuse inci- 
dent light and examine it specularly (Fig. Ic). 

The color quality of the light may influence the results and so may 
the color-sensitivity of the receiver. For example, a bluish tinted 
sample may have a high density to red light and a lower density to 
blue light. 

If photographic layers were neutral, non-scattering absorbers, it 
would not matter what geometrical optical system was used nor 
would the spectral quality of the components make much difference. 
Actually, they are neither. This fact makes it imperative that the 
pertinent conditions be thoroughly understood before a satisfactory 
densitometer can be designed. 

A moment's reflection will show that since the scattering power 
increases with an increased deposit of silver, the effect is to render 
specular densities higher than diffuse densities, and the density differ- 
ence increases with the absolute density. The ratio of specular den- 
sity D/> to diffuse density D# at a particular density level is known 
as the Callier quotient. In an effort to insure fundamentally accurate 
density measurements, instruments have recently been designed 
whose geometrical optics exactly conform to the concept of diffuse 



Feb., 1942] DIRECT-READING DENSITOMETER 151 

density. While such instruments are valuable in providing a readily 
accessible densitometer of the primary standard type, it will be 
shown later that conformity with theoretically ideal systems is quite 
unnecessary for densitometers used in routine work. After densi- 
tometers (using any practical optical system) have been calibrated 
as to diffuse density, they will give accurate readings on heterogeneous 
strips irrespective of grain size. 

The problem of color correction is not so simple. It is legitimate to 
hold that the effective spectral sensitivity of the optical system 
should duplicate practice. However, "practice" may mean almost 
anything. In the case of the motion picture industry, in the dupli- 
cating process alone an ideal system should duplicate several different 
combinations : 

(1) The combined spectral sensitivity of a mercury- vapor light- 
source or a tungsten light-source and dupe positive print material. 

(2) The combined spectral sensitivity of a mercury source and dupe 
negative. 

(5) The combined spectral sensitivity of a mercury source and re- 
lease positive. 

Different laboratories use different light-sources. Frayne 1 has met 
this problem by using a source-phototube-filter combination which 
approximates the spectral' sensitivity of a tungsten-positive-film com- 
bination. He further suggests that other combinations be adopted 
where special systems are to be duplicated. In the various branches 
of photography, nearly every conceivable type of optical system and 
color-sensitivity occurs. The result is that the effective density of even 
a neutral-density sample may vary by as much as 0.70 at a density 
level of 2.5. The selection of a representative criterion by which all 
densitometers might be calibrated has been discussed in the past, and 
references are cited which will acquaint readers with a few of the re- 
cent papers. 2 ' 3> * 5 In March of this year, an "American Recom- 
mended Practice for Photographic Density" (ASA, Z-22.27) was 
announced as follows : 

"The integrating sphere shall be used as a primary instrument for 
the determination of photographic density. Photographic densities 
determined by means of this primary instrument shall be used as 
secondary or reference standards by means of which densitometers of 
other types may be calibrated." 

This specification is somewhat inadequate. The spectral attributes 
of density just discussed are not even mentioned in the report. It is 



152 M. H. SWEET [J. S. M. P. E. 

the writer's opinion that some form of contact printing density would 
be found suitable for standardization. However, once any workable 
standard has been agreed on, definite coefficients could be assigned 
which would correct the standard values to agree with various optical 
systems. Additional coefficients could be calculated to help com- 
pensate for color differences. 

Requirements of the Densitometer. In the light of the above con- 
siderations, the following elemental requirements were formulated to 
create a goal toward which to work. 

(1) The instrument should be of the objective type. Visual densi- 
tometers impose severe eye fatigue if used continuously for long pe- 
riods of time. In addition, they are limited in accuracy by the con- 
trast sensitivity of the eye. 

(2) A ccuracy should be maintained to : 

0.005 from D = to Z) = 1.0 
0.010 from D = 1.0 to D = 2.0 
0.020 from D = 2.0 to D = 3.0 

These figures are derived from a study of the accuracy required for 
standard sensitometric procedures. 

(3) Stability should be commensurate with the specified accuracy. 

(4) It should be direct reading* and of the single-range type. This 
is to eliminate, so far as possible, the chance for mistakes on the part 
of the operator. 

(5) Good legibility should be maintained over the entire range of 
the instrument. Density readings should be made easily without im- 
posing eye-strain or physical discomfort. 

(6) The densitometer should be usable under ordinary room-light- 
ing conditions. This allows the instrument to be more conveniently 
located and permits working under more pleasant conditions. 

(7) It should be mechanically rugged and rapid in operation. 
Existing visual instruments are mechanically rugged, so the new 
instrument should be capable of withstanding limited misuse. Speed 
of operation is considered desirable in order to accelerate the sensi- 
tometric work. 

(8) It should read diffuse density. This aspect will be discussed 
separately. 

* Occasionally instruments of the balanced type are incorrectly referred to as 
direct-reading. True direct-reading instruments may be defined as those which 
indicate the density of the sample directly upon its insertion in the measuring beam 
without resort to further manual adjustments. 



Feb., 1942] 



DIRECT-READING DENSITOMETER 



153 



(9) The cost of the instrument as a whole should be competitive 
with that of visual instruments. Even if the other requirements were 
met, many prospective users would object to a cost appreciably higher 
than the prevailing cost of visual densitometers. 

(10) Further miscellaneous factors such as reasonable size, appear- 
ance, and permanence of calibration had to be considered. 

Having given careful attention to the factors outlined above, 
the model 11 densitometer was designed and constructed. 

Amplifier Ti/be 
Res is Tor- 
Photo Tube 



Fi/m 6dm pie 

Ah/o 



Condenser Lens 



L/Q ht source 



FIG. 2. Optical and mechanical pictorial diagram of model 11 direct-reading 

densitometer. 




DESCRIPTION OF DENSITOMETER 

Optical System. The present optical system is very simple. The 
optics are represented in Fig. 2. Light radiating from a 15-cp con- 
centrated-filament auto headlamp is focused by a single condenser 
on an aperture (not shown) . A heat-absorbing filter is used to elimi- 
nate the infrared radiation. The sample to be measured is placed im- 
mediately over the aperture which is located in the base plate of the 
instrument (Fig. 3). After passing through the sample, the beam 
is absorbed by the phototube. The phototube, together with the 
amplifier tube, is mounted in a hinged cylinder (Fig. 10). Orienta- 
tion of the sample with respect to the aperture is facilitated by raising 
the cylinder. 

From our discussion of density measurement, the present optical 
system will be recognized as being neither diffuse nor specular. It can 



154 M. H. SWEET \J. S. M. P. E. 

most satisfactorily be described as a double-diffuse system since the 
sample is diffusely illuminated and the emergent light is diffusely re- 
ceived. 

Referring to Fig. 4, a represents the path of the light-rays for a 
conventional optical arrangement. The density is, of course, mea- 
sured in terms of the flux absorbed by the photosensitive element 
with the sample interposed, in comparison with the flux absorbed 
without the sample. In Fig. 4(b) diffuse illumination is represented. 
Obviously, nearly all the light has a longer optical path through the 



POWER-SUPPLY 



HAXIMUM 

DENSITY 

ADJUSTMENT 



LIGHT SOURCE 
CONDENSING- LE.HS 



FILTER. 
APERTURE 




FIG. 3. View of underside of instrument showing location of components. 



photographic layer in the second case than in the first, with the re- 
sult that the apparent density of the sample is higher for diffuse illu- 
mination than for specular illumination. The magnitude of the dif- 
ference may be calculated from Beer's law. 

In an article on printing density, 12 Tuttle describes an apparatus 
whose optics are the equivalent of the components just described. 
This system gave higher density values than the simple diffuse 
type. However, Tuttle argued that the diffuseness of the source 
should not have influenced the results. 



Feb., 1942] 



DIRECT-READING DENSITOMETER 



155 



The model 11 densitometer was empirically calibrated, and the 
effect of the geometry of the illumination did not influence the ac- 
curacy of the values obtained. 

The full intensity of the beam falls on the phototube when zero den- 
sity is read. Otherwise, the intensity is modulated by the sample 
measured. The high sensitivity of the phototube-amplifier combina- 
tion makes it possible to use a low-candlepower lamp (15 cp) operated 

SAMPLE. 



COLUIMATED 
INCIDENT LIGHT^ 



PHOTOCELL 




DIFFUSE 
INCIDENT LIGHT 



(D) 




FIG. 4. Different optical systems for density measure- 
ment. 



at considerably less than normal voltage (5.0 volts). The life ex- 
pectancy of the lamp operated under these conditions is 40,000 hours. 
Color Response. The 929 phototube is the most sensitive vacuum 
phototube commercially available.* It is of the Sb-Cs coated type, 
and a plot of its spectral sensitivity is given in Fig. 5.** As mentioned 
earlier, Frayne 1 also used this tube and developed a source-phototube- 
filter combination which duplicates quite satisfactorily the color 



* Rated at 45 /^a/lumen for tungsten at 2870 K. 
** Adjusted for tungsten at 2670 K and 2-mm Corning No. 596 filter. 



156 



M. H. SWEET 



[J. S. M. P. E. 



attributes of an. optical system wherein positive film is the light-sen- 
sitive medium. Until the problem of standardization of density 
measurements has been definitely settled, it seems justifiable simply 
to use whatever color combinations seem most pertinent for each par- 
ticular case. In our meter, we are using the 15-cp source at a color- 
temperature of approximately 2670 K, the 929 phototube, and a 
2-mm light-shade Aklo heat-absorbing filter. 

Amplifier. Most direct-reading photoelectric instruments have 
used electrical circuits with essentially linear amplifier response to 
phototube output. If unmodified, this gives a badly cramped den- 
sity scale as shown in Fig. 6 (a). The scale characteristics of the bal- 



JDO 
80 
60 
40 
20 



I I 

929 PHOTOTUBE. + ^ 

C ORJM I N&- 396Fl LTER. 






E.YL 



\ 



JQQ w HO 60 BO^QQ 2O **0 60 80 2O 40 60 80 



100 
80 
60 
40 
20 



X 

FIG. 5. Relative response of the light -source-filter -phototube combination 
and the eye, to tungsten at 2670 K. 

anced-type densitometers depend on the particular optical scheme 
used and the design of the mechanical linkage. A combination can 
usually be found which will give a linear density scale. 

The uniformity of the scale of barrier-layer photocell-microammeter 
systems can be improved by inserting a high series resistance in the 
circuit. However, this requires higher initial light intensities, thus 
decreasing the inherent stability of the circuit. 

When using a photocell or phototube-amplifier combination whose 
output characteristics are linear with respect to intensity, it is feasible 
to use an output meter having cut pole-pieces. The Weston model 
877 meter 6 is an example of the photocell type and the ERPI model 
RA 1100 1 an example of the phototube-amplifier type. While the 
improvement is quite helpful, it by no means gives a uniform den- 



Feb., 1942] 



DIRECT-READING DENSITOMETER 



157 



sity scale. Furthermore, cut pole-piece instruments are not evenly 
damped over their whole scale lengths. Consequently, meters of this 
type are under-damped and/or over-damped at some part of their scale. 
These disadvantages may be avoided in a phototube-amplifier com- 
bination with a logarithmic response. Hardy 8 in 1929 described a 
logarithmic circuit for use in what is now called the GE Recording 
Spectrophotometer. Hardy's circuit is the prototype of the present 
design. Others 9 - 10 n since Hardy's original disclosure have described 

OENSITY 




DENSITY 

7 * * + 3 Z 



B \r 

FIG. 6. Densitometer scales: (-4) meter response proportional to 
incident intensity; (B) model 11 densitometer. 

variations of the same fundamental circuit. In general, these loga- 
rithmic circuits have been somewhat unsuccessful. They were too 
unstable, especially when required to cover an intensity range of 
1:1000 (this range of intensity corresponds to 0.0 to 3.0 density). 
Perfect linearity over this wide range was not claimed for any of these 
instruments. 

The theory of the present amplifier may be explained by referring 
to Fig. 7. In this illustration, the phototube is connected directly to 
the grid of the 6F5 amplifier tube (a conventional triode). Light 
falling on the phototube will create a grid current and the potential 



158 



M. H. SWEET 



[J. S. M. p. E. 



Output Meter 



ftCA-929 



of the grid will tend to become positive. This gives rise to an increased 
plate current which is measured by the output meter. Roughly 
speaking, the following relationships apply: 

(1) The light on the phototube is a linear (intensity) function of 
the transmission of the film sample. 

(2) The phototube current is a linear function of the incident light. 

(3) The phototube current = 
grid current. 

(4) The grid potential is a loga- 
rithmic function of the grid current. 

(5) The plate current is a linear 
function of the grid potential. 

Density is defined as D = log 
1/r, and step No. 4 introduces 
the logarithmic relationship nec- 
essary to give a uniform density 
scale on the output meter. The 
electronic theory which explains 
step No. 4 has been outlined by 
Russell. 9 

Fig. 8(^4) shows the relation- 
ship between sample density vs. 
plate current for the circuit of 
Fig. 7 (-4). Obviously it is not 
perfectly linear. In Fig. 7(B) a 
plate resistor has been added. 
This greatly improves the por- 
tion of the curve corresponding to 
the lower sample densities as 
shown in Fig. 8(B). 

In Fig. 7(C) a very high- value 
Its effect is to increase the cut-off 
value of the grid current of the tube as represented by the broken 
line of Fig. 8(C). By choosing the resistance value correctly, and by 
applying the proper bias voltage, a linear toe can be obtained. 

The action of this grid circuit modification is quite interesting. The 
grid resistor serves to introduce a current which opposes the photo- 
tube-grid current. The magnitude of this bucking current is only 
of the order of 0.01 microampere. At a high density of, say, 3.0, the 
phototube current is about 0.02 /ia and the bucking current has an 




FIG. 7. Development of linear 

density vs. output circuit: 
(.4) Fundamental circuit. 

(B) Plate resistor added. 

(C) Plate resistor and grid resistor 

circuit added. 

grid bias resistor has been added. 



Feb., 1942] 



DIRECT-READING DENSITOMET ER 



159 



/.o 

\ 

a 



Sample Density J 



appreciable influence on the grid potential. At a density of 2.0, the 
phototube current is 0.20 and the effect of the bucking current is very 
small. For densities near zero, the photocurrent is 20 ;ua and the 
effects of the bucking current are undetectable. Hence it is clear that 
this grid circuit arrangement operates in such a way as to affect the 
density vs. plate current curve 
only at high-density values. 

A schematic diagram of the 
circuit used in the model 11 
densitometer is shown in Fig. 9. 
It will be noted that a bucking 
current is used in the plate cir- 
cuit. This serves to shift the 
usable grid (and plate) current 
range to higher values giving 
better stability. 

Fig. 10 shows the measuring 
head unassembled. The photo- 
tube, grid resistor, and amplifier 
tube are mounted, end to end, in 
the cylinder. This affords ideal 
electrical shielding. It is un- 
necessary to shield the various 
leads to the cylinder since there 
are no high impedances, exclu- 
sive of the grid resistor, which 
are in any way delicate. The 
measuring head can be detached 
from the base of the densitome- 
ter. It is then useful as an "ex- 
ploring element" with which the 
intensity of light for small areas 






FIG. 8. 



D J 

Density vs. plate current for 
circuits of Fig. 5. 



of a projected image may be measured. The logarithmic response 
of the circuit has a unique advantage in this application because 
it is the logarithm of the relative light-intensity in the plane of the 
paper that is important in projection printing. 

The extreme sensitivity of the instrument permits the measure- 
ment of very high densities. With the present low-cp source, den- 
sities up to 5.0 have been measured. A 50-cp headlamp operated at 
7.5 volts should enable one to measure densities up to 6.0. 



160 



M. H. SWEET 



[J. S. M. P. E. 



Suitably mounted, the measuring head could be used in reflection 
densitometry. The advantages of the logarithmic circuit would apply 
to such an instrument. The excellence of linearity to logarithmic 




FIG. 9. Electrical circuit of model 11 direct-reading densitometer. 



WINDOW 




FIG. 10. Showing components of measuring head, unassembled. 

changes in light-intensity, together with the high speed of response 
of the circuit and the relatively high (1.0 ma) output render the cir- 
cuit adaptable for incorporation in the design of recording densi- 



Feb., 1942] 



DIRECT-READING DENSITOMETER 



161 



tometers. It should be especially valuable where continuous-tone 
sensitometer strips are to be automatically measured and recorded. 

Calibration. The hand-calibrated scale can, of course, be made to 
agree with any predetermined criterion. In the absence of wide- 
spread agreement on a standard type of density, the present densi- 
tometer was calibrated according to visual diffuse density. A care- 
fully prepared neutral density wedge was read repeatedly by different 
workers using a visual diffuse instrument. These values were used as 



ou 
28 
26 
24 
22 
<2 .20 
P .18 
$ 16 

g 14 
^ 12 

5 .ID 

| 

06 




































/ 


































/ 
































j 


? 
































f 


" 
































/\ 


































X 


































/ 


































X 
































^ 


/' 
































-^ 


^ 
































x> 


^ 
































,,/ 


^ 


























.04 

.Oi 

1 








^ 
































x 


^ 
































^x 


? 


































80 160 WO 12D 100 80 GO 40 2.0 C 

METER. READING- ( mm.) 



FIG. 11. Calibration of model 11 densitometer. 



cardinal points and the scale of the instrument was graduated so as 
to agree with them. 

The approximation to perfect linearity is illustrated by the fac- 
simile scale 0/2 size) shown in Fig. 6() . Fig. 1 1 is a plot of scale deflec- 
tion in mm s. visual diffuse density reading. Part of the apparent ir- 
regularities in the curve are known to be attributable to the uneven- 
ness of calibration of the visual densitometer which was used as the 
reference standard. 

To determine the influence of grain size on the calibration, a fine- 
grain positive sensitometer strip and a coarse-grain high-speed nega- 
tive were measured on both densitometers. Fig. 12 shows the results. 



162 



M. H. SWEET 



[j. a M. p. E. 



TABLE I 

Effect of Grain Size on Density Measurements 



Fine-Grain Positive Film 



Coarse-Grain Negative Material 



Visual 
Diffuse 
Density 


Photo- 
electric 
Density 


DHL 


Visual 
Diffuse 
Density 


Bhoto- 
electric 
Density 


Diff. 


0.05 


0.04 


-0.01 


0.18 


0.16 


-0.02 


0.09 


0.07 


-0.02 


0.21 


0.20 


-0.01 


0.13 


0.13 


0.0 


0.27 


0.25 


-0.02 


0.25 


0.23 


-0.02 


0.35 


0.34 


-0.01 


0.39 


0.37 


-0.02 


0.46 


0.42 


-0.04 


0.57 


0.56 


-0.01 


0.57 


0.54 


-0.03 


0.82 


0.82 


0.0 


0.70 


0.69 


-0.01 


1.10 


1.12 


+0.02 


0.86 


0.84 


-0.02 


1.40 


1.42 


+0.02 


1.00 


0.98 


-0.02 


1.71 


1.73 


+0.02 


1.13 


1.14 


+0.01 


2.02 


2.02 


0.0 


1.27 


1.29 


+0.02 


2.30 


2.30 


0.0 


1.41 


1.44 


+0.03 


2.54 


2.54 


0.0 


1.56 


1.56 


0.0 








1.69 


1.72 


+0.03 








1.84 


1.85 


+0.01 








1.99 


2.00 


+0.01 








2.12 


2.12 


0.0 








2.25 


2.26 


+0.01 








2.40 


2.40 


0.0 






TABLE 


n 






Effect 


of Staining 


Developer (Sease 3) on Density Measurements 


Visual 


Diffuse Density Photoelectric 


Density 


Difference 






0.41 


0.39 




-0.02 




/ 


0.43 


0.40 




-0.03 






0.46 


0.43 




-0.03 






0.53 


0.49 




-0.04 






0.61 


0.60 




-0.01 






0.70 


0.71 




+0.01 






0.83 


0.84 




+0.01 






0.97 


0.98 




+0.01 






1.07 


1.13 




+0.06 






1.22 


1.29 




+0.07 






1.33 


1.44 




+0.11 






1.51 


1.61 




+0.10 






1.68 


1.78 




+0.10 






1.85 


1.93 




+0.08 






1.96 


2.10 




+0.14 






2.12 


2.24 




+0.12 






2.22 


2.37 




+0.15 






2.28 


2.45 




+0.17 






2.35 


2.57 




+0.22 






2.43 


2.64 




+0.21 






2.53 


2.82 




+0.29 





Feb., 1942] 



DIRECT-READING DENSITOMETER 



163 



Visual diffuse density is represented on the horizontal axis and photo- 
electric density minus visual diffuse density is scaled on the vertical 
axis. Table I gives the corresponding numerical values. In neither 




"POSITIVE 



HlGHSPEEPNEGATlVE 
(COARSE &RAJN) 



.3 .4 .5 G 7 8 9 20 1- 2. 3 ** 

VISUAL DIFFUSE DENSITY 



FIG. 12. Correction curves for samples of different grain sizes. 




I 2. 3 



89UQ12.34-56 ' 8 

VISUAL DIFFUSE DENSITY 



20 L 



FIG. 13. Density difference (photoelectric vs. visual diffuse density) -for a 

stained film. 

case would it be possible in ordinary practice to discriminate between 
the photoelectric density and visual density curves for either of the 
two strips. 



164 



M. H. SWEET 



[J. S. M. P. E. 



The situation is more unfavorable in the case of stained films, and 
reddish colored samples may show much higher photoelectric densities 
than visual values. A plot similar to Fig. 12 is shown in Fig. 13 for 
a strip developed in Sease 3 developer for 30 minutes. Table II gives 
the numerical values. The correction curve is serious enough in this 
case to give different gammas and film speed determinations. 



.08] 
.06 



HP 



.06 



12.345- 



78910 




FIG. 14. Correction curve for a tinted base sample. 

TABLE III 

Effect of Tinted Base on Density Values 
Visual Diffuse Density* Photoelectric Density 



0.16 
0.20 

0.26 
0.28 
0.44 
0.86 



0.11 
0.14 
0.22 
0.24 
0.36 
0.78 



Difference 
-0.05 

-0.06 
-0.04 
-0.04 
-0.08 
-0.08 



Deposits on tinted bases are easier to deal with. For example, a 
blue base X-ray film gives good agreement with visual measure- 
ments when the base density is subtracted. A typical set of values 
is given in Table III, and the correction curve is shown in Fig. 14. 
Base-subtracted readings may be made directly once the light-in ten - 

* The visual diffuse densitometer used in this test was different from that used 
in all of the other tests. Its calibration is known to be somewhat different. 



Feb., 1942] DIRECT-READING DENSITOMETER 165 

sity has been increased so as to give a zero reading with a section of 
the base alone in position. 

The permanence of calibration of the instrument, both from day to 
day and in terms of months has been studied. Daily checks have 
been made since the instrument was first installed. Table IV is largely 
self-explanatory. It gives the readings made with the densitometer 
at a few typical daily and monthly intervals. At all densities up to 
2.0, the variation in readings was negligible. At densities beyond 2.5 
there was a noticeable tendency to give slightly higher readings with 
time. It is believed that this change is due to aging of the grid re- 
sistor, although it is possible that the amplifier tube characteristics 
may have changed. In any event, the shift is not serious, and can be 
corrected once every week or two by readjusting the grid bias poten- 
tial. 

TABLE IV 

Permanence of Calibration of Densitometer* 



Calibra- 
tion 














Wedge 


Original 


17 Days 


18 Days 








Step No. 


Readings 


Later 


Later 


19 Days 


20 Days 


25 Days 


1 


0.07 


0.07 


0.07 


0.07 


0.07 


0.08 


2 


0.09 


0.09 


0.09 


0.09 


0.09 


0.10 


3 


0.15 


0.15 


0.15 


0.15 


0.15 


0.15 


4 


0.26 


0.26 - 


0.26 


0.25 


0.25 


0.26 


5 


0.43 


0.43 


0.43 


0.42 


0.42 


0.43 


6 


0.67 


0.67 


0.67 


0.66 


0.66 


0.67 


7 


0.94 


0.93 


0.94 


0.93 


0.93 


0.94 


8 


1.22 


1.21 


1.21 


1.20 


1.20 


1.21 


9 


1.45 


1.45 


1.45 


1.42 


1.43 


1.45 


10 


1.66 


1.67 


1.67 


1.66 


1.67 


1.68 


11 


1.88 


1.89 


1.89 


1.89 


1.88 


1.90 


12 


2.10 


2.10 


2.11 


2.10 


2.09 


2.11 


13 


2.30 


2.31 


2.31 


2.31 


2.30 


2.33 


14 


2.49 


2.50 


2.50 


2.51 


2.51 


2.53 


15 


2.67 


2.69 


2.69 


2.71 


2.69 


2.71 


16 


2.85 


2.88 


2.88 


2.90 


2.90 


2.92 



It is important to consider the effect of changing amplifier tubes. 
If their characteristics were not alike, the meter scale would have had 
to be recalibrated. To explore this possibility, a second 6F5 was se- 
lected at random and used to replace the original. Table V shows the 

* The grid bias was left unadjusted during the period covering the above test. 
This was done to show the drift in high-density readings. Later, by decreasing 
the grid bias, ^he readings were made to agree with the original values. 



166 M. H. SWEET [J. S. M. P. E. 

TABLE V 

Effect of Replacement of Amplifier Tube* 



Tube No. 1 


Tube No. 2 
Installed 43 Days Later 


Difference 


0.07 


0.07 


0.00 


0.09 


0.09 


0.00 


0.15 


0.15 


0.00 


0.26 


0.26 


0.00 


0.43 


0.44 


+0.01 


0.67 


0.68 


+0.01 


0.94 


0.95 


+0.01 


1.22 


1.22 


0.00 


1.45 


1.45 


0.00 


1.66 


1.67 


+0.01 


1.88 


1.88 


0.00 


2.10 


2.09 


-0.01 


2.30 


2.28 


-0.02 


2.49 


2.48 


-0.01 


2.67 


2.66 


-0.01 


2.85 


2.83 


-0.02 



* Note that the grid bias adjustment was imperfect as evidenced by the dis- 
agreement at high densities. 



TABLE VI 

Effect of Warm- Up Period on Meter Readings 





True 


After 












Density* 


1 Min 


3 Min 


5 Min 


10 Min 


15 Min 


1 


0.07 


0.08 


0.075 


0.075 


0.075 


0.075 


2 


0.09 


0.11 


0.095 


0.09 


0.09 


0.095 


3 


0.15 


0.17 


0.16 


0.15 


0.15 


0.155 


4 


0.27 


0.28 


0.26 


0.27 


0.27 


0.27 


5 


0.44 


0.43 


0.44 


0.44 


0.44 


0.44 


6 


0.69 


0.69 


0.69 


0.69 


0.69 


0.69 


7 


0.95 


0.95 


0.96 


0.96 


0.96 


0.95 


8 


1.22 


1.23 


1.23 


1.23 


1.23 


1.22 


9 


1.45 


1.44 


1.45 


1.46 


1.45 


1.45 


10 


1.67 


1.67 


1.66 


1.68 


1.68 


1.67 


11 


1.88 


1.88 


1.88 


1.89 


1.88 


1.88 


12 


2.07 


2.10 


2.10 


2.10 


2.09 


2.10 


13 


2.28 


2.30 


2.30 


2.30 


2.30 


2.30 


14 


2.48 


2.50 


2.49 


2.49 


2.49 


2.49 


15 


2.66 


2.68 


2.67 


2.67 


2.68 


2.68 


16 


2.86 


2.87 


2.84 


2.87 


2.86 


2.88 



* Readings made 16 hours earlier after the instrument had been thoroughly 
warmed up. 



Feb., 1942] DIRECT-READING DENSITOMETER 167 

comparison. It is obvious that a recalibration is unnecessary. A 
maximum density difference of 0.02 exists over a very limited region. 
Otherwise the agreement is within 0.01. 

Performance. Recalling the requirements set forth in the intro- 
duction, it will be realized that the first few, including accuracy, 
legibility, etc., have been met. An inexpensive, low-power voltage- 
regulator stabilizes the power-supply voltages. The lamp voltage 
is taken from the filament winding of the secondary coil and is 
trimmed by a variable resistor. This gives stability quite consistent 
with the original specifications. Table VI demonstrates the warm-up 
period. One minute after turning the densitometer "on" (it having 
been left unoperated for 16 hours), a set of readings was taken; three 



ZERO ADJUSTMENT 




err* -SWITCH 



FIG. 15. Showing appearance of the model 11 densitometer in use. 

minutes later, a second set; five minutes later, a third set, etc. Evi- 
dently the warm-up characteristics primarily affect only the higher 
density values. An idea of the size and appearance of the finished in- 
strument may be gained from Fig. 15. 

To show the freedom from zero drift, readings were made under 
conditions identical with those just described. The instrument was 
turned "on" and allowed to warm up for one minute. The zero ad- 
justment was made. Every minute thereafter, for five minutes, the 
zero adjustment was repeated and a record made of the displacement 
of the pointer from zero. These values are shown in Table VII. 
After the instrument had been in operation for twenty minutes, a 
zero reading was made every 5 seconds for a total of one minute. 
At no time did the pointer vary by the width of the (knife-edge 



168 M. H. SWEET [J. S. M. P. E. 

type) pointer. This corresponds to a fluctuation of approximately 
two thousandths density. 

TABLE VII 

Effect of Warm- Up Period on Zero Drift 

After 1 minute of operation zero was reset 

2 minutes of operation correction was +0.005 

3 minutes of operation correction was +0.002 

4 minutes of operation correction was . 000 

5 minutes of operation correction was . 000 

6 minutes of operation correction was . 000 

The influence of line voltage changes has been studied. After a 
thorough warm-up, the supply voltage was varied in steps of 5 volts, 
from 95 to 125 volts, and the zero readings were recorded. The data 
for this test are compiled in Table VIII. Apparently a variable sup- 
ply voltage is not a problem. 

TABLE Vm 

Influence of Line Voltage 
Line Voltage Zero Reading 

95 +0.005 

100 0.000 

105 (normal) 0.000 

110 -0.002 

115 -0.010 

120 -0.010 

125 -0.010 

Ease of reading on the part of the operator has been insured by us- 
ing an output meter of excellent legibility. The scale length is 178 
millimeters, about 7 inches. (The ordinary 3-inch panel instrument 
has a scale length of only 50 mm.) This long scale is essential to an 
accurate single-rate instrument. It allows the divisions to be spaced 
iy 5 mm apart, each division corresponding to 0.02 density. In 
searching for practical objections to the densitometer, operators 
have been asked whether or not the legibility was satisfactory. All 
agreed that a still longer scale would not further improve the ease 
of reading. 

The output meter has a sensitivity of 1.0 ma. It is not at all slug- 
gish, having a speed of response of about */& second, and is uniformly 
damped over its entire range. Its steel case minimizes the effects 



Feb., 1942] DIRECT-READING DENSITOMETER 169 

of external magnetic and electrostatic fields besides protecting it 
from mechanical damage. 

In our laboratories, trained operators require approximately 45 
minutes to read and record the densities of ten Type lib sensitometer 
strips using conventional visual instruments. With the model 11 
densitometer, this time is reduced to only 20 minutes. 

All the component parts of the instrument are rigidly mounted on 
the Y^-inch thick base-plate. The plate is flush-mounted in a small 
table built for the purpose. This makes servicing very convenient, 
for the whole instrument may be removed as a unit to a work-table 
and the various parts are then readily accessible. 

TABLE IX 

Results Obtained by Different Observers Using the Model 11 Densitometer 

Step Observer 

No. No. 1 No. 2 No. 3 No. 4 No. 5 No. 6 

1 0.06 0.05 0.05 0.05 0.06 0.05 

2 0.085 0.08 0.08 0.09 0.085 0.08 

3 0.14 0.14 0.13 0.15 0.14 0.12 

4 0.28 0.28 0.28 0.29 0.29 0.28 



5 


0.49 


0.48 


0.48 


0.49 


0.49 


0.48 


6 


0.79 


0.79 


0.79 


0.79 


0.785 


0.79 


7 


1.13 


1.13 


1.12 


1.13 


1.13 


1.12 


8 


1.43 


1.43. 


1.43 


1.43 


1.44 


1.43 


9 


1.68 


1.67 


1.68 


1.68 


1.69 


1.68 


10 


1.90 


1.89 


1.90 


1.90 


1.91 


1.90 


11 


2.09 


2.08 


2.09 


2.10 


2.10 


2.09 


12 


2.25 


2.25 


2.26 


2.26 


2.26 


2.25 


13 


2.39 


2.39 


2.39 


2.40 


2.40 


2.39 


14 


2.52 


2.51 


2.51 


2.52 


2.53 


2.52 


15 


2.65 


2.66 


2.66 


2.67 


2.67 


2.68 



16 2.79 2.79 2.78 2.80 2.81 2.80 

As a final test, six individuals were asked to read the same sensi- 
tometer strip on the model 11 densitometer. Some were previously 
unfamiliar with the instrument. The results are compiled in Table 
IX. Of the 96 values tabulated, only two lie outside the specified ac- 
curacy limits (readings 4 and 6 of step No. 3). Even in the case of 
these readings, the errors may be partly attributed to unevenness in 
the density of the sample. The measuring aperture is about 3 mm in 
diameter and shifting the position of ordinary samples within a given 
step will produce density variations up to 0.03 (due in part to the 
Eberhard effect) . Furthermore, the presence of dust particles on the 



170 M. H. SWEET [J. S. M. P. E. 

sample will sometimes increase the apparent density. Nevertheless, 
the accompanying data are considered sufficiently reliable to demon- 
strate the performance of the instrument as it is used in actual prac- 
tice. 



CONCLUSIONS 

A simple direct-reading densitometer has been built which facili- 
tates the measurement of transmission densities. The electronic 
arrangement was so designed as to give a strictly uniform density 
scale over the entire range covered (0.0 to 3.0). It has been demon- 
strated that for the simple optical system used, the accuracy of the 
instrument was unaffected by the grain size of the materials mea- 
sured. 

Reproducibility, stability, drift, and other factors were shown to 
have a negligible influence on the accuracy of the instrument. In 
general, the densitometer was shown to fulfill the requirements agreed 
on before its construction was undertaken. 



REFERENCES 

1 FRAYNE, J, G.: "Measurement of Photographic Printing Density," J. Soc. 
Mot. Pict. Eng., XXXVI (June, 1941), p. 622. 

2 TUTTLE, C., AND KoERNER, A. i "Standardization of Photographic Density," 
/. Soc. Mot. Pict. Eng., XXIX (Dec., 1937), p. 622. 

8 "American Recommended Practice for Photographic Density ASA,Z-22.27," 
J. Soc. Mot. Pict. Eng., XXXVI (March, 1941), p. 245. 

4 KOERNER, A., AND TUTTLE, C.: "Experimental Determination of Photo- 
graphic Density," J. Opt. Soc. Amer., XXVII (July, 1937), p. 241. 

6 SWEET, M. H.: "Photographic Density Determined by the Contact Print- 
ing Method," /. Opt. Soc. Amer., VH (Oct., 1941), p. 126. 

6 ROCKWELL, H. P., JR : "Exposure Makes the Print," Bull., Weston Elec- 
trical Instrument Corp., Newark, N. J. (1939), p. 9. 

7 FRAYNE, J. G., AND CRANE, G. R.: "A Precision Integrating-Sphere Densi- 
tometer," /. Soc. Mot. Pict. Eng., XXXV (Aug., 1940), p. 184. 

8 HARDY, A. C.: "A Recording Photoelectric Color Analyzer," /. Opt. Soc. 
Amer., XVIII (1929), p. 106. 

9 RUSSELL, J.: "A Photoelectric Cell Circuit with a Logarithmic Response," 
Rev. Sci. Instr., VIII (Dec., 1937), p. 495. 

10 MILLER, C. W.: A Linear Photoelectric Densitometer," Rev. Sci. Instr., VI 
(Apr., 1935), p. 125. 

11 TIEDMAN, J. : "A Linear Densitometer," Electronics (March, 1941), p. 48. 

" TUTTLE, C.: "Densitometry and Photographic Printing: Illumination of 
the Negative and Its Effect on Density," J. Opt. Soc. Amer. (Oct., 1934), p. 272. 



Feb., 1942] DIRECT-READING DENSITOMETER 171 

DISCUSSION 

DR. FRAYNE: The author stated in his paper that one of the reasons for under- 
taking the development of this densitometer by his organization was the fact that 
there was no commercial instrument of this type available. I wish to point out 
that the precision integrating sphere densitometer developed by Electrical Re- 
search Products, Inc., and described in the August, 1940, issue of the JOURNAL has 
been made commercially available during the year 1941 and has had wide adop- 
tion in the motion picture laboratories and studios in Hollywood and elsewhere. 
To indicate the almost universal adoption of this instrument at this time, I would 
like to show its disposition in the various organizations. 

Number of 
Instruments 

20th Century-Fox West Coast Lab. 3 

M. G. M. Laboratory 3 

Paramount West Coast Laboratory 2 

Paramount East Coast Laboratory 2 

Consolidated West Coast Laboratory 1 

De Luxe Laboratory 1 

Pathe West Coast Laboratory 1 

Technicolor 1 

Williams Laboratory 1 

Triplett & Barton X-Ray Laboratory 1 

DuPont Film Manufacturing Co. 1 

Eastman Kodak (Rochester) 1 



These instruments are not being used primarily as primary standards in these 
various institutions, but are in active use in daily control operations. It is ob- 
vious, from the number and the distribution of the instruments that have been 
made available to the motion picture industry within one year, that this instru- 
ment most certainly may be regarded as being commercially available. 

The operation of the instrument is extremely simple and measurements can be 
made rapidly and accurately by relatively unskilled observers. The multi-range 
feature permits an extended scale for each density range and adds materially to 
the ease of making observations. 

MR. SWEET: I did not intend to disparage the merits of the ERPI densi- 
tometer which I sincerely believe to be an excellent piece of equipment. Never- 
theless, this instrument does not meet the essential requirements set forth in the 
introduction of our paper, and, furthermore, it suffers from some of the objec- 
tions which apply to many photoelectric instruments. 

(A) The ERPI densitometer is certainly not simple, and complexity was given 
as a criticism of most objective instruments. The electrical circuit consists of a 
multi-stage amplifier with inverse feedback. Its optical system is relatively com- 
plex, using a chopping disk, diffusing block, sphere, etc. These complications are 
always a potential source of trouble. Furthermore, the amplifier requires several 
balance and zero adjustments for regular operation. 

(B) The cost of the ERPI sphere instrument is considered prohibitive by 
many laboratories, at least for application involving routine work. Its cost is in 
excess of $1000, whereas the present instrument costs a fraction of this figure. 



172 M. H. SWEET 

(C) While the ERPI densitometer may be considered a direct-reading instru- 
ment it is of the multi-range type, and this feature was explained as being unde- 
sirable. The present instrument is of the single-range type, and still affords ex- 
cellent legibility over the entire density range covered. 

As a standard instrument, Dr. Frayne's densitometer may establish itself in the 
photographic industry but the writer feels he has adequately demonstrated that a 
much simpler instrument of the type described will suffice for routine work. 

MR. CARLSON: What is the influence of line voltage changes on the accuracy 
of the meter readings? 

MR. SWEET: The effect is very small. Table VII shows the magnitude of the 
shift in the zero reading for a line- voltage variation of 95 to 125 volts in increments 
of 5 volts. A Sola 30-watt magnetic-type regulator is used in this instrument be- 
cause in the immediate application the voltage fluctuations were unusually severe. 
The regulator is obviously very effective in preventing any sizable zero drift. 
In applications where normal fluctuations exist it might be feasible to dispense 
with the regulator. 



AN ANALYSIS OF THE APPLICATION OF FLUORESCENT 
LAMPS TO MOTION PICTURE PHOTOGRAPHY* 



REINHARDT ROSENBERG* 



Summary. A brief discussion on photographically important characteristics 
of fluorescent lamps is followed by a thorough mathematical analysis of the constancy 
of light produced by fluorescent lamps. The conclusion is reached that, during ex- 
posure times ordinarily encountered in motion picture photography, the constancy of 
light from fluorescent lamps, if these are properly applied, far exceeds minimum 
requirements and that, even for extremely short exposure times, this constancy is 
well within accepted limits. 

The readiness with which still photographers have accepted fluores- 
cent lamps as illuminants invites an investigation of their use in 
commercial motion picture photography. The high luminous ef- 
ficiency and the lack of heat and glare of these lamps would seem to 
make them highly desirable as light-sources, as their use would effect 
considerable savings in operating expense and would, also, greatly 
increase the comfort of technicians and actors. However, it must be 
left to writers more competent on these subjects to weigh the savings 
in operation against .the initial investment, and to determine whether 
the low surface brightness of fluorescent lamps is compatible with 
the high levels of illumination prevalent in modern motion picture 
studios. It is intended to limit this article to a careful analysis of 
the physical possibility of using these lamps for motion picture photog- 
raphy, and to the photographic and technical aspects involved. 

Only the 3500 white, the blue, and the daylight fluorescent lamps 
can be considered for general illumination. The others are too dis- 
tinctly "colored" to be of use for photographic purposes, and the 
light of some is filtered which reduces their luminous efficiency. As 
has been shown elsewhere 1 the three lamps mentioned above differ 
in the following characteristics: (1) luminous efficiency, i. e., pro- 
duction of lumens per watt; (2) photographic efficiency, i. e., the 

* Received October 18, 1941. 

** Department of Anatomy, University of Pittsburgh School of Medicine, 
Pittsburgh, Penna. 

173 



174 



R. ROSENBERG 



[J. S. M. P. E. 



shortest exposure time required to produce a, given density under 
standard conditions and at a specified gamma; (3) energy distribu- 
tion over the photographically effective range of the spectrum; (4) 
flicker, or time-energy characteristic; and (5) shape of the .D-log E 
curve of an emulsion exposed to the various lamps. 

Each lamp, compared with either of the other two, is superior in 
some characteristics and inferior in others. The white lamp possesses 
the highest luminous efficiency and the least flicker. However, its 
photographic efficiency is less than that of either the blue or the day- 
light lamps. The flicker of this lamp is only slightly less than that 
of the daylight lamp. Furthermore, the D-log E curve of a pan- 





WAVELENGTH IN A 

FIG. 1. Spectrogram of the daylight fluorescent lamp, photo- 
graphed on a Wratten tricolor panchromatic plate. 

chromatic emulsion, when exposed to the white lamp, has a shorter 
straight-line portion than is the case when exposed to the blue or the 
daylight lamp. It is, therefore, not a good choice. 

Of the three lamps, the blue lamp has the highest photographic 
efficiency, and exposure of photographic emulsions to it produces a 
straighter D-log E curve than either the daylight or the white 
lamp. It can not, however, be considered a good choice as it has by 
far the worst flicker and the worst spectral characteristics. Its 
energy in the red portion of the spectrum is negligible so that it pro- 
duces on panchromatic materials results similar to those of an or- 
thochromatic emulsion. It would therefore require radical and un- 
desirable changes in make-up technic. 

By the process of elimination the daylight fluorescent lamp must 
be considered the proper choice. All its characteristics, when con- 



Feb., 1942] 



FLUORESCENT LAMPS 



175 



sidered separately, are inferior to either the 3500 white or the blue 
lamp, but taken as a group they indicate that this lamp is the most 
desirable of the three. The spectrogram of the daylight lamp, photo- 
graphed on a type B panchromatic emulsion, is shown in Fig. 1. 
This illustration is similar to the well known wedge spectrograms 
and its interpretation is subject to the same limitations. Fig. 2 
shows a typical series of D-log E curves and a time-gamma curve, 
also obtained from a type B panchromatic emulsion, and Fig. 3, 
curve A, illustrates the nicker or time-energy characteristic of the 
daylight fluorescent lamp. Curve A in Fig. 3 was obtained by photo- 




L ,E.po(AH,i ( ,.-Sc.U) 



FIG. 2. D-log E curves and time-gamma curve of 
Eastman portrait panchromatic film, exposed to the 
daylight fluorescent lamp. 

graphing the screen of a cathode-ray oscilloscope which was tracing 
the current induced in a photoelectric cell when exposed to a day- 
light fluorescent lamp, operated on 60-cycle, 110- volt, alternating 
current. The most serious objection to the use of this or any other 
fluorescent lamp for illumination in motion picture photography is 
the fluctuating character of the light it produces. In order to explain 
this flicker a knowledge of the following facts is helpful : 

Fluorescent lamps are glass tubes whose walls are covered on the 
inside with fluorescing substances. A low-pressure mercury arc 
within the tube produces ultraviolet light which, in turn, excites the 
fluorescence. As in all gas-discharge lamps, little energy is stored 
in the fluorescent lamp, and, twice during each cycle, when the cur- 



176 



R. ROSENBERG 



[J. S. M. P. E. 



rent goes through zero, the production of ultraviolet light ceases. 
The lamp does not become completely dark, however, since fluores- 
cence persists to some extent until the next impulse occurs. Al- 
though the flicker can be eliminated by operating the lamp on direct 
current, this is not advised by the manufacturer because the lamp 
does not operate as satisfactorily on direct as on alternating current, 
and because its high luminous efficiency is largely sacrificed when it is 

operated on direct current. The 
effect of the flicker can, how- 
ever, be overcome if three banks 
of equal numbers of fluorescent 
lamps are operated on each of the 
three phases of three-phase al- 
ternating current. 

The constancy of light so ob- 
tained will now be analyzed. In 
this analysis the ripple on top of 
the wave, apparently caused by 
a secondary emission, will be 
disregarded since its magnitude 
is negligible compared with the 




N+.^SECOND N 

T,=TIME 

FIG. 3. Curve A shows the time- 
energy characteristic of a daylight 
fluorescent lamp, operated on 60- 
cycle, 110-volt alternating current. 
Curve B is the graphical presenta- 
tion of the function of curve A, 
harmonically analyzed to close con- 
gruence. For the sake of pictorial 
clarity curve B is shifted horizontally 
to the right. 



total light output and its fre- 
quency is too high to be of prac- 
tical significance. Curve A, Fig. 

3, was harmonically analyzed by making use of Fourier's series of 

the well known form 



= A + A l sin Z + A, sin '~\ + ... A n sin "' 



where 



cos + B z cos 2- + . . . B n cos - 



Ao = i 'ydx 

a Jo J 



Feb., 1942] FLUORESCENT LAMPS 177 

The analysis was carried to the fifth harmonic, resulting in the func- 
tion 

E l = /(/O = 691.6 -f 12 sin (0.1744/0 + 2.45 sin (0.3495/0 + 0.34 sin (0.523/0 
+ 0.72 sin (0.6975/0 + 0.25 sin (0.872/0 - 3.77 cos (0.1744/0 
4- 1.43 cos (0.3495/0 + 0.15 cos (0.6975/0 + 0.45 cos (0.872/0 

where 

EI = instantaneous luminous energy 

/i may assume any values between and 1 /i2o second 

Vm second = 2ir = 36 cm in the original curve. 

The curve of this function is shown in curve B, Fig. 3 which is suf- 
ficiently congruent with curve A for all practical considerations. 
The position of EI = was calculated from the manufacturer's data 
concerning the variation from the mean of the instantaneous lumen 
output of this lamp. 2 

The instantaneous lumen output of three lamps, operated out of 
phase on three-phase alternating current is given by the function 

3 = /(/) = /(/O + f(ti + Vaeo) + /(*i + VIM) 

and is shown graphically in/(/), Fig. 4. 

Since the function E 3 = f(t) is not a constant it is apparent that 
the amounts of light produced during a given time-interval will vary 
with the position of this interval in relation to the period of the func- 
tion and that they will vary less as the interval becomes longer. If, 
during the exposure- time per frame, these amounts should vary by 
more than 10 per cent, a standard suggested elsewhere in the litera- 
ture, 3 fluorescent lamps would have to be rejected as illuminants for 
motion picture photography. 

The maximum and minimum amounts of light produced during 
1 /48 second will now be calculated. This interval was chosen since 
it is obtained with the common camera speed of 24 frames per second 
and a shutter opening of 180 degrees. All possible amounts of light 
produced during 1 /4 8 second are expressed by the integral 



f" + 1/ -** = *(.) = f" +V "l/W* - f" + V " I/O,) + f(k + Vae.) + 
Jn Jn Jn 

f (/i + VMO)] dt 



where n may assume any values between and Vseo second. Twelve 
integrations of this function were carried out, assuming for n succes- 

sive values of - , - , - , . . . - . These inte- 
12 X 360 11 X 360 10 X 360 360 



178 



R. ROSENBERG 



[J. S. M. P. E. 




grations revealed that the luminous output of three daylight fluores- 
cent lamps, operated under conditions indicated above, and taken 
during random intervals of V second, does not vary by more than 
0.1 per cent, since the calculations, which were carried to the third 
significant number, showed no variations at. all. 

It is easily understood that amounts of light emitted during a 

given interval will show greater varia- 
tions as the interval becomes shorter, 
reaching the greatest possible varia- 
tion at infinitesimally short intervals. 
Shortened exposure times occur in 
three instances: (1) in fades and lap 
dissolves made by reducing the shutter 
opening; (2) in slow-motion photog- 
raphy; and (3) in full exposures with 
reduced shutter opening. Although it 
is the most frequent, the fade is per- 
haps the least important of these since 
it lasts only a short time, since exposure 
times per frame vary continuously 
during the operation, and since a poor 
fade will hardly detract from the 
merits of an otherwise good motion 
picture. Slow-motion photography is 
usually applied only in outdoor sport 
events or, when light conditions are 
not under the control of the motion 
picture operator, but it is worthy of 
consideration since it might find ap- 
plication in studio trick photography. 
The most important instance of 
shortened exposure time is, un- 
doubtedly, full exposure with reduced 
shutter opening since this is an important device to improve the 
definition of an image that moves with great rapidity across the image 
plane. 

For these instances of reduced exposure time it might be inter- 
esting either to find the shortest exposure during which a predeter- 
mined constancy of illumination can be maintained or, assuming 
an infinitesimally short exposure, to calculate the maximum and 



TIME IN -^SECOND 

FIG. 4. Curves /&), f(ti + 
1/360), and /(/i + 1/180) show, 
in proper time relationship to 
each other, the time-energy 
curves of three daylight fluores- 
cent lamps, operated on three- 
phase current. Curve /(f) illus- 
trates the time-energy curve of 
the light produced by the com- 
bination of the three lamps, i. e., 
their algebraic sum. 



Feb., 1942] FLUORESCENT LAMPS 179 

minimum values for E z . This latter was carried out by letting the 
first derivative of the function E 3 = f(t) equal zero : 

dEt/dt = /'(/) = 

and, thereby, determining the values of / at which the maximum 
and minimum occur. The values for t, so obtained, were then sub- 
stituted in the equation E s = f(t) and the magnitudes of the maximum 
and the minimum were determined. It was found that, during in- 
finitesimally short exposures, the least amount of light is 88.9 per 
cent of the maximum or, in other words, the variation from the mean 
is only about 5 per cent. It is reasonable to deduce that, for the 
shortest exposure times practicable, the previously accepted standard 
of a 10 per cent variation is easily maintained. 

In conclusion, it can be stated that daylight fluorescent lamps, if 
properly applied, can be successfully used as illuminants in motion 
picture photography, i. e., they can produce illumination of great 
constancy, and their light is photographically satisfactory. In 
addition, their use may result in considerable economies. 

Grateful acknowledgment is made to the General Electric Com- 
pany for making available the lamps and fixture used in this work. 

REFERENCES 

1 ROSENBERG, R.: "Film Characteristics with Fluorescent Lamps," Photo 
Technique (May, 1941), p. 50. 

2 "Engineering Data on Fluorescent Lamps," General Electric Company. 

3 MOLE, P. : "A New Development in Carbon Arc Lighting," J. Soc. Mot. Pict. 
Eng., XII (Jan., 1934), p. 51. 



IODIDE ANALYSIS IN AN MQ DEVELOPER* 

R. M. EVANS, W. T. HANSON, JR., AND P. K. GLASOE** 



Summary. A method is described for the analysis of iodide in a developer in- 
volving precipitation of the halide with silver nitrate and oxidation of the iodide while 
it is in the form of solid silver iodide to iodate with chlorine water. The iodate is then 
determined polar o graphically. Quantities of iodide from 2.5 to 10 milligrams of po- 
tassium iodide were analyzed with an accuracy of 2-4 per cent. Thiocyanate in the 
developer interferes but it can be removed by boiling with strong sulfuric acid before 
precipitation. 

Using this method of analysis it was shown that an equilibrium amount of iodide is 
obtained in a developer. Curves are given showing the attainment of equilibrium for 
development of Eastman panchromatic negative motion picture film, in Kodak D-76, 
and in Kodak D-16 developers, and Eastman positive motion picture film in Kodak 
D-16. The equilibrium value depends on the emulsion, the developer, the developed 
density, and perhaps other variables which will be investigated more thoroughly. 

During the last several years, an attempt has been made to apply 
chemical analysis to the complete control of photographic developers. 
Evans and Hanson, 1 Baumbach, 2 and Atkinson and Shaner 3 have 
published analytical methods for most of the important constituents 
of developers. 

One result of a complete knowledge of the chemical composition of 
an aged developer would be the ability to match the developing prop- 
erties of a used developer by means of a freshly mixed solution. Re- 
cent attempts to accomplish this by using only the analyses already 
published have not given entirely satisfactory results. There are 
some indications that the amount of iodide which accumulates in a 
used developer plays a significant role in the developing properties 
and must be added to the synthesized formula. Because of this 
need, the following method for determining iodide in a used developer 
solution has been worked out. 

Rylich 4 first demonstrated that iodate ion could be reduced at the 
dropping mercury electrode. This reaction is the basis of the test. 

* Received May 15, 1941; communication No. 808 from the Kodak Research 
Laboratories. 

** Eastman Kodak Company, Rochester, N. Y. 
180 



IODIDE IN AN MQ DEVELOPER 181 

The dropping mercury electrode 5 is composed of mercury dropping 
from a capillary tube immersed in a solution. The mercury column 
within the capillary is connected externally to a pool of mercury at 
the bottom of the container (or to a calomel electrode which is im- 
mersed in the solution) by means of a variable source of potential and 
a galvanometer. When a potential difference is applied, a current 
flows. This current fluctuates, reaching a maximum just before each 
drop falls, then decreasing suddenly when the drop falls and slowly 
rising to the maximum again. As this potential is increased, the 
maximum current increases slowly. However, if in the solution some 
material is present which is reducible at the dropping mercury elec- 
trode, it will begin to be reduced at a potential characteristic of this 
material and a large increase in the current will occur. As the po- 
tential is increased further, the current- voltage curve flattens off. 
This gives rise to a "wave," the height of which is proportional to the 
amount of the material present. This height can be used as a method 
for determining the amount of an electroreducible material present. 

In order to determine iodide in a developer, the iodide, which is not 
electroreducible, must be oxidized to iodate, and interfering sub- 
stances which are reducible at or near the reducing potential of iodate 
must be removed. Separating the iodide by precipitation was 
avoided at first in the hope of keeping the procedure as simple as pos- 
sible. Several attempts were made to oxidize iodide to iodate di- 
rectly in a developer solution. To eliminate the necessity for using 
excessive amounts of the oxidizing agents, the sulfite was removed by 
acidification and by passing steam through the solution, and the de- 
veloper was removed by extraction with ethyl acetate. Nevertheless, 
a large amount of some oxidation products which gave high currents 
was always formed and interfered with the results. Several differ- 
ent oxidizing agents, both solid and dissolved, were investigated. 
Attempts were also made to remove the undesirable oxidation prod- 
ucts by extraction and adsorption but without success. 

The next attempts were made using weaker oxidizing agents that 
would only oxidize the iodide to iodine which could be extracted with 
carbon tetrachloride and later oxidized to iodate. Using this pro- 
cedure, only a small fraction of the iodide present could be accounted 
for, and it was concluded that the iodine formed by oxidizing the 
iodide reacted with the oxidation product of the developer so was not 
available for extraction by the carbon tetrachloride. 

After these failures it was concluded that the iodide would have to 



182 



EVANS, HANSON, AND GLASOE 



[J. S. M. P. E. 



be separated from the developer solution by precipitation with silver. 
In this way all of the halide present in the developer, including the 
iodide, could be separated from the interfering substances in the de- 
veloper by decantation and washing. At first, it was considered nec- 
essary to put the precipitated silver halide into solution before oxi- 
dizing the iodide to iodate. This again produced interfering sub- 
stances and led to a rather involved method for removing them. 

It was finally found that the pro- 
cedure could be shortened consider- 
ably by oxidizing the iodide in the 
precipitate itself rather than after 
redissolving. This eliminated all of 
the interfering substances and yielded 
curves as good as those obtained 
from solutions of pure potassium 
iodate. 

In order completely to oxidize the 
iodide in the precipitate, an oxidiz- 
ing agent must be used which is 
sufficiently powerful to oxidize the 
bromide also, thereby exposing any 
iodide in the interior of a particle to 
the action of the oxidizing agent. 
Chlorine water was found to be quite 
satisfactory. This gives complete 
oxidation of the iodide and leaves 
behind an exceedingly fine precipi- 
tate of silver chloride which can be 
removed by filtration. The filtra- 
tion is made more complete by the 
addition of a small amount of kieselguhr which coagulates the silver 
chloride and prevents it from passing through the filter paper. 

The next step is the removal of the excess chlorine by the addition 
of phenol. This reacts immediately with the chlorine but does not 
react with iodate, and an excess of phenol has no influence on the 
curves. 

The potential at which the curve for iodate occurs is a fairly steep 
function of pH in acid solutions but is independent of pH in alkaline 
solutions. For this reason the solution containing the iodate is made 
alkaline. 



60 



55 



50 



40 



20 



15 



10 




0.9 1.0 I.I 1.2 1.3 1.4 1.5 1.6 
VOLTS 

FIG. 1. Dropping mercury 

electrode "waves" for iodate; 

amount of iodate equivalent to 

(Curve 1) 2 l /2 mg/1 of KI 

(Curve 2} 5 mg/1 of KI 

(Curve 3} 7 l /o mg/1 of KI 

(Curve 4} 10 mg/1 of KI 



Feb., 1942] 



IODIDE IN AN MQ DEVELOPER 



183 



The instrument used in these Laboratories for obtaining the curves 
is the Fisher Elecdropode.* A simple potentiometer-galvanometer 
set-up could be used equally satisfactorily. The usual precautions 
necessary in such work must be observed. 5 The solution being an- 
alyzed must be bubbled with nitrogen for several minutes before elec- 
trolysis to remove oxygen, and it must be held at constant tempera- 
ture during the reading. 

In calibrating the test with solutions containing a known amount 
of iodide, it is desirable to obtain the entire wave for iodate by reading 



5 o 

if" 

tt 
o 

z 



5 10 

CONC. OF KI IN THE DEVELOPER (M<j/L) 

FIG. 2. Calibration curve derived from Fig. 1. 

the current at a series of potentials. Such curves are shown in Fig. 
1. The wave is completed between 0.9 and 1.4 volts, using a 
saturated calomel electrode. If, now, the difference in current be- 
tween these two voltages is plotted against the known concentration 
of potassium iodide in the developer, a straight line is obtained, as 
shown in Fig. 2. Once these correct voltages have been established, 
an unknown can be determined merely by reading the current at 
these two points and determining the concentration from the calibra- 
tion curve. 



* Fisher Scientific Co., Pittsburgh, Pa. Recent Developments in Laboratory 
Appliances, Fisher Elecdropode, Cat. No. 9-317. 



184 



EVANS, HANSON, AND GLASOE 



[J. S. M. P. E. 



The exact procedure for the analysis is as follows : 

Take a 100-cc sample of developer. 

Add 10 cc of 0.5 N potassium bromide and 60 cc of cone, sulfuric acid (to insure 
sufficient precipitate to coagulate). 

Allow steam to pass through the solution or boil it until evolution of gas ceases. 

To the hot solution add 100 cc of water and 25 cc of 0.5 N silver nitrate and pass 
steam through solution or boil it for a few seconds to aid in coagulation of the pre- 
cipitate. 




10 20 30 40 50 60 10 80 90 100 110 120 
FEET OF FILM PER LITER 

FIG. 3. Iodide equilibrium for Eastman panchromatic 
negative motion picture film developed for 15 minutes in 
Kodak D-76 to a density of 1.0. 



Allow the precipitate to settle and pour off the supernatant liquid, leaving the 
precipitate in the flask. 

To the precipitate add 50 cc of 1 : 1 nitric acid and 250 cc of water; shake up 
precipitate and allow to settle. 

Pour off the supernatant liquid. 

Repeat this process with two 250-cc portions of water. 

To the precipitate add 50 cc of fresh chlorine water. 

Heat until bromine color disappears. 

Filter and wash the precipitate with two 5-cc portions of water. 

To filtrate add 1.0 cc of 5% phenol,, 2 drops of phenolphthalein and enough 
2.0 N potassium hydroxide to give a pink color (4-10 drops). 



Feb., 1942] 



IODIDE IN AN MQ DEVELOPER 



185 



Dilute to 100 cc with pH 10.0 buffer (sodium carbonate, 4 grams per liter; 
borax, 4.7 grams per liter). 

Bubble nitrogen through the solution, check temperature, and electrolyze with 
dropping mercury electrode. 

Read concentration from calibration curve. 

The accuracy obtainable is shown in Table I in which the results 
of the analyses of developers containing known amounts of potassium 
iodide are tabulated. 




O 10 20 30 -40 SO 60 10 80 90 100 HO 120 
FEET OF FILM PER LITER 

FIG. 4. Iodide equilibrium for Eastman pan- 
chromatic negative motion picture film developed 
for 5 minutes in a positive developer to a density of 
1.5. 



For developers containing potassium thiocyanate, the procedure 
just described must be altered slightly. In the normal procedure 
the thiocyanate is precipitated by the silver and interferes with the 
curve for iodate. However, the thiocyanate is destroyed slowly by 
sulfuric acid and if, after the acidification and passing steam through 
the solution, the developer is allowed to stand for two or three min- 
utes and steam is passed through it again, the thiocyanate is removed 



186 EVANS, HANSON, AND GLASOE [J. S. M. p. E. 

completely. If steam is not readily available, the solution may 
simply be boiled to give the same result. 

Several different types of MQ developers containing known 
amounts of potassium iodide have been analyzed and all gave satis- 
factory results on the same calibration curve. Consequently, the 
test need not be calibrated for each formula used, provided no new 
compounds are present. However, the presence of materials other 
than the normal constituents of an MQ developer, as, for example, po- 
tassium thiocyanate, may necessitate some alteration in the procedure. 

TABLE I 

Potassium Iodide Added Potassium Iodide Found 

(milligrams per liter) (milligrams per liter) 

2.5 2.6 

5.0 5.0 

7.5 7.5 

10.0 9.9 

5.0 Sample 1 4.9 

2 4.9 

3 4.8 



This procedure has been used to determine the iodide equilibrium 
which is established in developers during film aging. As expected, 
this equilibrium is a function of the developer formula, developing 
time, density developed, type of emulsion, and possibly other vari- 
ables. These are still being investigated, but the preliminary results 
are worthy of some comment. 

Eastman panchromatic negative motion picture film was exposed 
uniformly in a printer so that it gave a density of about 1.0 in 15 min- 
utes of development in Kodak D-76 developer. At several footage 
intervals, samples of the developer were analyzed for potassium io- 
dide. The results are shown in Fig. 3 curve a, where the amount of 
potassium iodide in a liter is plotted against the number of feet of film 
developed per liter. After about 20 feet per liter, the iodide remains 
constant at about 3 milligrams per liter. Bromide analyses made at 
the same intervals are shown in curve b. To make sure that a true 
equilibrium had been reached, a quantity of potassium iodide greater 
than the equilibrium amount was added to the developer and the 
procedure repeated. The results are shown in curve c. The iodide 
decreases and approaches the same equilibrium that was obtained 
previously. 



Feb., 1942] 



IODIDE IN AN MQ DEVELOPER 



187 



In Fig. 4 the iodide equilibrium for Eastman panchromatic film 
developed for 5 minutes to a density of 1.5 in a positive-type developer 
is shown to be about 15 milligrams per liter. This demonstrates 
that, for a given emulsion developed to a given density, the equilib- 
rium is a function of the developer formula. The equilibrium for 
Eastman positive motion picture 
film developed for 5 minutes to 
a density of 0.5 in a positive de- 
veloper is shown in Fig. 5 to be 
about 1 milligram per liter. 

Including iodide as one of the 
important constituents, analyses 
of a few used developers have led 
to synthesized formulas which 
checked photographically with 
the used developers. This work 
is being continued in the hope 
that in the near future it will be 
possible to predict the photo- 
graphic activity of a developer solution directly from analyses. 



i 

Of 




10 20 30 40 50 60 10 80 90 



FIG. 5. Iodide equilibrium for 
Eastman positive motion picture 
film developed for 5 minutes in a 
positive developer to a density of 0.5. 



REFERENCES 

1 EVANS, R. M., and HANSON, W. T., JR. : "Chemical Analysis of an MQ De- 
veloper," /. Soc. Mot. Pict. Eng., XXXII (March, 1939), p. 307. 

2 BAUMBACH, H. L.: "The Chemical Analysis of Metol, Hydroquinone, and 
Bromide in a Photographic Developer," /. Soc. Mot. Pict. Eng., XXXIII (Nov., 
1939), p. 517. 

3 ATKINSON, R. B., AND SHANER, V. C.: "Chemical Analysis of Photographic 
Developers and Fixing Baths," J. Soc. Mot. Pict. Eng., XXXTV (May, 1940), p. 
485. 

4 RYLICH, A.: "Polarographic Studies with the Dropping Mercury Cathode: 
Electroreduction and Estimation of Bromates and lodates," Collections Czech. 
Chem. Comm., 7, (1935), p. 288. 

6 KOLTHOFF, I. M., AND LiNGANE, J. J. : "The Fundamental Principles and 
Applications of Electrolysis with the Dropping Mercury Electrode and Hey- 
rovsky's Polarographic Method of Chemical Analysis," Chem. Rev., 24, (Feb., 
1939), p. 1. 



SYNTHETIC AGED DEVELOPERS BY ANALYSIS* 
R. M. EVANS, W. T. HANSON, JR., AND P. K. GLASOE** 

Summary. The dropping mercury electrode is applied to the problem of analyzing 
aged photographic developers and new tests are described for elon and hydroquinone. 
The question of suitable tests for bromide is discussed and it is shown that the bromide 
test must be independent of chloride. Such a test is described. 

Using these new technics and others it is demonstrated that it is possible and practi- 
cal by chemical analysis alone to match exactly the photographic characteristics of an 
aged MQ developer. The only elements necessary for such an analysis are elon, hy- 
droquinone, sulfite, salt concentration, pH, bromide, and iodide. The precision re- 
quired for the proper analysis for each constituent has been investigated and is reported 
for three developer-film combinations. In general, the precision required is different 
for every combination. 

It is pointed out that the results show there are no unknowns in aged developers, 
that they differ from the original formulas only in concentrations and in the presence 
of iodide as well as bromide. Several useful consequences are suggested. 

(I) INTRODUCTION 

For the past several years, work has been carried out in these labo- 
ratories on the chemical analysis of photographic developers with 
the hope of being able to predict the action of a photographic de- 
veloper from its chemical analysis. It is the opinion of many workers 
that an aged developer is in some way different in its photographic 
action from a fresh developer made up to correspond to the analysis 
of the aged one. The data given in this paper show that this differ- 
ence does not exist under the conditions of these experiments. De- 
velopers aged with motion picture positive or negative film have been 
analyzed and the photographic action of these developers matched by 
a fresh developer made up to the formula analyzed. 

However, before this could be accomplished, it was necessary to 
make a study of the sensitometric changes caused by a given differ- 
ence in developer formula in order to know how accurate the analysis 
had to be. Having determined the required precision of analysis, the 

* Presented at the 1941 Fall Meeting at New York, N. Y. ; received October 
22, 1941. Communication No. 826 from the Kodak Research Laboratories, 
** Eastman Kodak Company, Rochester, N. Y, 

18$ 



SYNTHETIC AGED DEVELOPERS 189 

analytical methods available were examined. If the method was 
found to give results in error by more than the allowable amount, 
attempts were made to improve the existing method or a new one was 
worked out. 

(n) EXAMINATION OF DEVELOPER VARIABLES 

(A) Experimental. Two emulsions were used : Eastman motion 
picture positive and panchromatic motion picture negative. Sensito- 
metric exposures were made on a type lib sensitometer. The ex- 
posure on positive film was made with a positive lamp calibrated for 
a color-temperature of 2660 K with a light blue filter and a gelatin 
neutral density of 0.3. This gave log E max = 1.75. The exposure 
on negative film was made with a negative lamp calibrated for a 
color-temperature of 2360K with a No. 79 filter and a gelatin neu- 
tral density of 0.6. This gave log E max = T.92. 

The processing equipment used for these tests consisted of a stain- 
less-steel developing machine provided with removable 4-liter tanks 
and a separate pair of 25-liter stainless-steel washing tanks. 

The developing machine was similar to that described by Jones, 
Russell, and Beacham 1 but on a smaller scale. It was equipped to 
agitate six solutions at a time by means of flat paddles permanently 
attached to the individual tanks and guided both at the top and bot- 
tom. Each tank was fitted to hold two film-racks with the paddle 
passing between them. The paddles were 5 /s inch wide and passed 
about */4 inch from the emulsion surface at a speed of 8 inches per 
second, completing 25 cycles in one minute. 

The racks were frames arranged to hold the strips of film on indi- 
vidual slides. The individual slides were designed to take 12-inch 
strips with two holes punched at each end, holding them under light 
spring tension to facilitate loading and to take care of any expansion 
of the film on wetting. Each rack held four slides. 

The darkroom in which the sensitometric work was done was main- 
tained at 70F and 50 per cent relative humidity. The developing 
tanks were surrounded by water maintained at 68F == 0.1 F by 
means of a thermostat. 

In the washing tanks a strong flow of water and vigorous air agita- 
tion were combined to give sufficient washing conditions for twelve 
racks at a time. After thorough washing, the strips were dried in a 
forced draft drying cabinet operated at a standard temperature of 



190 EVANS, HANSON, AND GLASOE [J. S. M. P. E. 

All sensitometric strips were read with a recording physical densi- 
tometer. Four strips were developed on each rack and the curves 
given are the average of these four strips. 

Two types of developer were used, a D-16 type positive developer 
and a D-76 type negative developer. Positive strips were developed 
in positive and negative developers; negative strips were developed 
in the negative developer. Four development times were used with 
each set of developers. With the negative developer the times were 
4, 8, 16, and 32 minutes. With the positive developer the times were 
1, 2, 4, and 8 minutes. 

TABLE I 

Negative Film in Negative Developer 

Approx. Value in Aged 
Variable Developer Per Cent Error Tolerable 

Elon 1.5 6 

Hydroquinone 4.0 50 

KBr 1.0 4-5 

KI 2.0 mg 10 

pH 8.4 0.02Hunit 

Positive Film in Negative Developer 
Elon 1.5 3 

Hydroquinone 4.0 10 

KBr 1.0 4 

KI 2.0 mg 10 

pH 8.4 0.02 pH unit 

Positive Film in Positive Developer 
Elon 1.0 10 

Hydroquinone 3.0 10 

KBr 1.0 10 

KI 1.0 mg 20 

pU 10.00 0. 05 pH unit 

A sample of negative developer was aged with negative emulsion 
which had been given a uniform exposure to give a density of ap- 
proximately 1.0. This aged developer was analyzed to establish the 
order of magnitude of each of the constituents after such treatment. 
Then a series of four developers was made up in which one of the 
analyzed materials, for instance, elon, was varied in concentration 
above and below the value found on analysis, holding the other 
concentrations at the analyzed value. Strips of negative emulsion 
were developed in these developers, and from the set of sensitometric 



Feb., 1942] SYNTHETIC AGED DEVELOPERS 191 

curves so obtained it was possible to determine how much the elon 
concentration must change to give a density difference of 0.02 when 
the amount of elon present is approximately that found in the aged 
developer. The smallest increment of elon concentration which would 
cause this effect was used to determine the accuracy needed in the 
analytical test. Using the formula of the aged developer as a basis, 
this procedure was repeated for hydroquinone, bromide, iodide, and 
pH. Previous experience indicated that the other constituents of the 
developer could be neglected. 

The above procedure was repeated, using positive film strips in- 
stead of negative film strips in negative developer and in positive de- 
veloper. The data so obtained give the analytical precision necessary 
to give a check when positive film is developed in negative developer 
aged with negative film and in positive developer aged with positive 
film. 

The approximate percentage errors which could be allowed in each 
case are given in Table I. 

(B) Discussion. It is evident from these results that the vari- 
ables must be much more closely controlled in negative developer 
than in positive. pH, elon, and bromide are the variables which 
must be most accurately measured. Hydroquinone is relatively un- 
important and iodide can vary by a large percentage although the 
amounts involved are very small. 

(Ill) METHODS OF ANALYSIS 

A method of analysis for constituents of a photographic developer 
must be simple enough to be readily adapted to routine control work 
and yet accurate enough to be within the limits which will give a 
photographic effect. Such a method for hydroquinone in an MQ de- 
veloper has been in use in these laboratories for some time. 2 This 
test involves the separation of hydroquinone from elon by extraction 
with ethyl acetate from an acid solution followed by oxidation of the 
hydroquinone by potassium ferricyanide. The intensity of the 
brown color so produced is determined on an opacimeter and the con- 
centration of hydroquinone is obtained from a calibration curve. A 
test for elon was available but the method was rather involved and 
the results were not trustworthy. 

(.4) Elon Test. The method of separation of elon from a de- 
veloper solution has been worked out by Pinnow 3 and Lehmann and 
Tausch. 4 The procedure involves the extraction of elon with an or- 



192 



EVANS, HANSON, AND GLASOE 



[J. S. M. p. E. 



ganic solvent at the pH. at which elon is the least soluble in water and 
most soluble in the solvent. However, the methods available for the 
determination of the amount of elon in the organic solvent have been 
difficult and time consuming. The success of the test described here 
lies in the use of a new method of determining the concentration of 
elon in the extracting solvent. The polarograph is used to make such 
a determination rapidly and accurately. Since hydroquinone can 



O.IO O.OO OIO 0.2O 



POTENTIM. (VOLTS) 

FIG. 1. Polarographic curve for elon in pSL 8.0 
buffer. 



also be determined with the same technic, a simple test for hydro- 
quinone was worked out at the same time. 

A detailed description of the polarograph and its applications is 
given in the papers by Kolthoff and Lingane 5 and by Muller. 6 The 
instrument used in this work was the Fisher Elecdropode. When the 
polarograph is used in the determination of elon and hydroquinone, 
the dropping mercury electrode is the anode and the substance being 
determined is oxidized at the electrode, the current obtained being an 
anodic current. A typical polarographic wave for elon is shown in 



Feb., 1942] SYNTHETIC AGED DEVELOPERS 193 

Fig. 1. Since elon and hydroquinone are oxidized at very nearly the 
same potential on the polarograph, it is necessary to separate them 
before making the measurement. 

The method of separation involves the extraction of hydroquinone 
with ethyl acetate from an acid solution in which the elon is present 
as a salt and therefore is insoluble in the organic solvent. The sub- 
sequent extraction of elon is accomplished after the pH. of the solution 
has been adjusted to a value of approximately 8.0. The complete 
procedure for elon is as follows : 

To a 25-cc sample of developer in a 250-cc separatory funnel, add a 
few drops of bromphenol blue and 50% sulfuric acid until the color 
changes to yellow. Add 1.0 cc of this acid in excess. Add 50 cc of 
ethyl acetate and shake vigorously for a short time. Allow the lay- 
ers to separate, draw off the water layer, and repeat the process. 
After the second extraction, draw off the water layer, add to it a few 
drops of 8.4 indicator (one part 0.1% cresol red, three parts 0.1% 
thymol blue) and enough 2.0 N caustic to obtain a pink- violet color. 
Dilute the solution to 50-cc with pH 8.0 buffer* and extract the elon 
with 25 cc of ethyl acetate. Add 6 cc of the ethyl acetate layer to a 
mixture of 30 cc of 95% alcohol and 30 cc of H 8.0 buffer on the 
polarograph which has had the oxygen removed by bubbling nitrogen 
through the solution for about five minutes. Record the reading of 
the galvanometer at a potential of +0.1 volt against the saturated 
calomel cell. The concentration of elon is then found by reference to 
a calibration curve for which the data are given in Table II. The 
galvanometer readings given were obtained at a sensitivity of two 
times on the Elecdropode. The amount of the ethyl acetate extract 
to be used will be governed by the amount of elon in the sample. 
The amounts given above gave nearly full-scale deflection for a con- 
centration of 2.0 grams per liter in the original sample. This gave a 
maximum sensitivity for the concentration range 0-2.0 grams per 
liter. In a different concentration range the amounts used should be 
adjusted to give approximately full-scale deflection for the top con- 
centration. With this method an analysis for elon could be made in 
about ten minutes with an error of less than 5 per cent. Usually the 

* pH buffer formula: 

Boric acid 9.9 grams 

Potassium hydroxide 0.9 gram 

Water to make 1000 cc 



194 EVANS, HANSON, AND GLASOE [J. S. M. P. E. 

results checked within 2-3 per cent. Using this procedure, an exami- 
nation was made of the factors affecting the extraction of elon. 

TABLE II 
G/L Elon Galvanometer Reading at = +0.10 Volt 

0.0 2.5 

0.5 23.5 

1.0 45.5 

1.5 67.0 

2.0 88.5 

The acidity of the solution at the time of the hydroquinone extrac- 
tion was found to be very important. If the solution was titrated 
just to the bromphenol blue end point (/>H = 3.5), a small amount of 
the elon was extracted with the hydroquinone. The addition of 1 cc 
excess acid made the pH low enough so that no elon was removed. 
This fact was established by using a developer solution containing no 
hydroquinone. The ethyl acetate extract after the first acidification 
was tested for elon. A definite polarographic wave was obtained 
when the solution was titrated to pH 3.5 and no wave was obtained 
when the 1 cc excess acid was added. Table III shows the results 
obtained on a positive type developer with and without the excess 
acid. 

It was found necessary to extract the hydroquinone with two sepa- 
rate portions of ethyl acetate. If only one extraction was performed, 
some hydroquinone was left in the solution which was extracted at a 
pH of 8.0 and gave a polarographic wave when the concentration of 
elon was zero. With the two extractions, the amount of hydroqui- 
none was reduced enough so that no wave was obtained in the elon test 
when no elon was present. 

TABLE III 

G/L Elon G/L Elon 

Acidity Added Found 

Bromphenol blue end point 2.0 1 . 35 

Bromphenol blue end point 2.0 1 . 36 

Bromphenol blue end point 2.0 1 . 20 

1 cc excess acid over end point 2.0 2 . 05 

1 cc excess acid over end point 2.0 1 .97 

1 cc excess acid over end point 2.0 2.02 

To determine the effect of pH on the extraction of elon, different 
amounts of potassium hydroxide were added to the acid solution from 



Feb., 1942] SYNTHETIC AGED DEVELOPERS 195 

which the hydroquinone had been extracted. The amounts of alkali 
were such that one solution was just at the indicator end point, one 
had been titrated to the end point and then acid added to give the 
original yellow, and the third had obviously gone over the end point. 
The results are given in Table IV. It is apparent from these results 
that the pH of the solution from which the elon is extracted can vary 
from the value 8.0 much more than would ever be encountered in ac- 
tual practice, using the indicator given, without vitiating the result. 

TABLE IV 
G/L Elon Added pH G/L Elon Found 

2.0 8.48 2.00 

2.0 7.75 2.04 

2.0 9.10 1.99 

After making the solution alkaline, it is diluted to a definite volume 
because it was found that changing the volumes changed the amount 
of elon extracted. When a dilution was made to 50 cc, the concentra- 
tion of elon found was 2.0 grams per liter. When the dilution was 
made to 100 cc, the value obtained was 1.88 grams per liter. 

If the test is to be used on aged developers, the results must not be 
affected by the sulfonates of elon and hydroquinone. Since these sub- 
stances are present as salts at a H of 8.0, no extraction by an organic 
solvent is to be expected. Two samples of developer were made up, 
each containing 2.0 grams per liter of elon. The one containing no 
sulfonate gave an analysis of 2.0 grams per liter of elon. The other 
sample containing 8.0 grams per liter of hydroquinone monosulfonate 
analyzed 1.98 grams per liter of elon. 

In routine work it is inconvenient to have a different calibration 
curve for each different type of developer. It has been shown by 
Baumbach 7 that the per cent extraction of hydroquinone depends on 
the sulfite concentration and it follows that the same effect would be 
found in the case of elon. 

Developers containing varying amounts of sulfite and carbonate 
were analyzed for elon to determine the magnitude of this effect in 
this method of analysis. The results are given in Table V. The cali- 
bration curve was made on a solution containing 45 grams per liter 
of sodium sulfite and 30 grams of sodium carbonate. The column 
represented as grams per liter of sodium sulf ate gives the calculated 
amount of sodium sulfate equivalent to the sodium sulfite and 
sodium carbonate present. 



196 EVANS, HANSON, AND GLASOE [j. S. M. p. E. 

TABLE v 

G/L Elon 
G/L Sodium G/L Sodium G/L Sodium 

Sulfite Carbonate Sulfate Added Found 

20 0.0 22.5 2.00 1.90 

20 8.9 32.5 2.00 1.90 

40 0.0 45. Q 2.00 1.99 

40 26.7 75.0 2.00 1.99 

45 30.0 85.3 2.00 2.00 

80 0.0 90.2 2.00 2.05 

100 0.0 112.0 2.00 2.05 

100 40.0 145.0 2.00 2.07 

80 62.3 160.2 2.00 2.07 

These results indicate that although an increase in salt concentra- 
tion increases the amount of elon extracted, the same calibration 
curve could be used for salt concentrations equivalent to 45 to 160 
grams per liter of sodium sulfate. 

Some difficulty was experienced at first in getting a sharp break in 
the current-voltage curve obtained with an ethyl acetate extract. 
At that time the electrolyzed solution was buffered to pR 10. It was 
found that if the pH of the solution was 8.0 instead of 10, the curve ob- 
tained had a good break. The alcohol in the solution is necessary to 
make the ethyl acetate miscible with the buffer mixture. Sodium 
sulfite or any halides will interfere with the polarographic wave. 
For this reason, the calomel cell saturated with chloride is connected 
to the electrolysis cell by means of an agar bridge saturated with po- 
tassium nitrate. Also, one must be very careful not to introduce any 
of the water solution containing sulfite into the electrolyzed solution. 

(B) Hydroquinone Test. A polarographic wave similar to that of 
elon can be obtained with hydroquinone in the same buffer-alcohol 
mixture used in the elon test. However, if the ethyl acetate solution 
of hydroquinone obtained in the extraction from an acidified devel- 
oper is added to the buffer-alcohol mixture, no wave is obtained. 
This is due to the fact that the ethyl acetate dissolves a large amount 
of sulfur dioxide from the acid solution which then forms sulfite in 
the alkaline mixture. The presence of this sulfite will eliminate the 
break in the current-voltage curve. This sulfur dioxide can be re- 
moved by washing the ethyl acetate extract with an alkaline solution. 
The ethyl acetate solution so treated is added to the buffer-alcohol 
mixture and a good polarographic wave is obtained. 

The procedure used is as follows: To a 25-cc sample of developer 
in a 250-ml separatory funnel, add a few drops of bromphenol blue 



Feb., 1042] 



SYNTHETIC AGED DEVELOPERS 



197 



and then add 50% sulfuric acid until the color changes to yellow. Add 
1 cc of the acid in excess. Add 50 cc of ethyl acetate and shake vigor- 
ously for a few moments. Allow the layers to separate and draw off 
the water layer which will be used for an elon analysis if it is desired. 
To the ethyl acetate solution add 50 cc of a solution composed of 100 
grams sodium sulfite, 10 grams boric acid, and 1.0 gram potassium 
hydroxide in one liter. Shake up this mixture well and allow the 
layers to separate. Add 2 cc of the ethyl acetate extract to a mixture 
of 30 cc 95% alcohol and 30 cc of pH 8.0 buffer through which nitro- 
gen has been bubbled for about five minutes. The current at +0.10 
volt against the saturated calomel electrode is recorded and the con- 
centration of hydroquinone obtained from a calibration curve pre- 
viously made. 

The effect of a change in total salt concentration is less in the case 
of hydroquinone than it was in the case of elon. Table VI shows the 
results of analyses for hydroquinone in developers of varying salt 
content. 



Concentra- Concentra- 
tion of So- tion of So- 
dium Sulfite dium Carbon- 
G/L ate G/L 



10 

100 
100 







80 



TABLE VI 

Concentra- 
tion of So- 
dium Sulfate 
G/L 

11.2 
112.0 
203.0 



Concentration of Hydroquinone 
Added Found 



5.0 
5.0 
5.0 



5.05 

5.0 

4.9 



The first ethyl acetate extraction of the acidified developer re- 
moves approximately 92 per cent of the hydroquinone. Baumbach 7 
has given the value of extraction of hydroquinone with ethyl ether as 
60 per cent. This work indicates that ethyl acetate is a more effective 
extraction solvent. The ethyl acetate extraction of elon removes only 
about 80 per cent of the elon present. This difference in extraction 
coefficient explains the greater effect of changing salt concentration 
on the analytical result in the case of elon. 

When these two tests are performed on the same 25 cc of developer 
they can be completed in 15 to 20 minutes with an accuracy of 5 per 
cent or better. Before a calibration curve is made, a few trial analyses 
should be performed in which the whole current- voltage curve is 
plotted and a point about midway in the flat portion of the curve 
chosen as the potential at which readings will be made. 



198 



EVANS, HANSON, AND GLASOE 



[J. S. M. P. E. 



(C) Bromide Analysis. The test for bromide 2 which involved SL 
titration with an adsorption indicator was considered to be sufficiently 
accurate and it was used in this work for some time. However, as 
Evans and Hanson 2 pointed out, this method will also titrate chloride 
with the bromide so that if any chloride is present, the bromide analy- 
sis will be high. Using a silver electrode, a potentiometric titration 
of an aged developer solution which had been acidified gave two 



0060 

0030 

0.0 

0.030 

00*0 

0.090 

OJ20 

0.150 

0.180 

o.zio 

^0*40 

0.270 

|0300- 

0.330 




4.0 6.0 fcO 
CC. Aq N0 4 O.05N 



ao 



FIG. 2. Potentiometric titration of film aged 
developer silver electrode. 

breaks (Fig. 2) indicating the presence of chloride as well as bromide. 
Although the amount of chloride is small, there is enough to make the 
bromide analysis in error by 10-15 per cent. 

The photographic influence of the presence of chloride in the two 
developers used was investigated over the range of concentrations 
from to 8 grams per liter and it was found to have no effect. How- 
ever, the sensitivity to bromide was such that if the bromide analysis 
included tine chloride so that in the synthetic developer the chloride was 
replaced by an equivalent amount of bromide, an appreciable error 
would result. Fig. 3 shows the photographic result obtained in such a 



Feb., 1942] 



SYNTHETIC AGED DEVELOPERS 



199 



case. By the adsorption indicator method of analysis the concentra- 
tion of potassium bromide found was 1.4 grams per liter while the 
value obtained by the potentiometric titration method was 1.07 
grams per liter. This finding explained the failure to obtain matching 
sensitometric curves for aged and fresh developers, using the adsorp- 
tion indicator method of analysis. 

The analytical method for bromide in an aged developer which was 
finally used is as follows: To 25 cc of developer, which has been 



2.7 
2.0 
1.6 
I* 

1.4 

> in 



0.8 
0.6 
0.4 
O.Z 




12 15 

LOG E 
16 MINUTES 



18 21 



21 30 



FIG. 3. Negative film in negative developer. 
Fresh developer 
O Aged developer 
A Synthetic aged developer 
D Synthetic aged developer without iodide 
X Synthetic aged developer. Bromide analysis by ad- 
sorption indicator. 



heated to boiling to complete the reduction of any silver held in solu- 
tion by the sulfite and then cooled,* add 10 cc of 50% sulfuric acid. 

Pass steam through the solution or boil it to drive off the dissolved 
sulfur dioxide. 

Add 20 cc of a solution of sodium acetate, 150 grams per liter, to the 
above solution and titrate the solution with 0.057V silver nitrate, using 
a silver electrode and a regular potentiometric set-up for potentio- 



* This step is necessary only with low-energy high-solvent developers such as 
D76. 



200 



EVANS, HANSON, AND GLASOE 



[J. S. M. P. E. 



metric titrations.* A plot of the points will readily show the break 
due to bromide. This method requires more time than the indicator 
method, but with a little experience an analysis can be made in about 
fifteen minutes. One gram of potassium bromide per liter can be de- 
termined with an accuracy of two per cent. 



2.4 




24 




2.2 




2.2 




2.0 




2.0 




1.8 




1.8 




> | g 




>l.6 


_ 


S 1.4 
uil.2 




w|.4 
gl.2 


x x 


1.0 


^ 


I.O 




0.8 


^^" 


0.8 




0.6 


^.'' 


0.6 


'*' ^^ 


0.4 


^-"'\'_^^~^ 


0.4 

r\o 


'^^^'^^v-'^*' 


0.2 


^^^^^^^^^^^^ , 


O.2 




C 


) 3 <o 9 12 15 18 21 24 27 3( 


) ( 


) 3 <b 9 12 15 18 21 24 27 3C 



LOG E 
4 MINUTES 



LOG E 
8 MINUTES 




16 MINUTES 



15 18 21 24 27 30 
LOG E 
32 MINUTES 



FIG. 4. Negative film in negative developer. 

Fresh developer 

O Aged developer 

A Synthetic aged developer 



(D) pH. The pH of the developer was measured with a Beck- 
man Laboratory Model pR meter. Since in matching an aged de- 
veloper with a fresh one it was not important to have the absolute 

* In this work the Fisher Elecdropode was used by simply substituting the 
silver electrode for the dropping mercury electrode and inserting a tap key in the 
circuit so that the galvanometer could be Used as a null-point instrument. 



Feb., 1942] 



SYNTHETIC AGED DEVELOPERS 



201 



value of pR but merely to have the same pH, the sodium-ion correc- 
tions for the glass electrode were neglected. It was found to be pos- 
sible to get the two developers to the same H within 0.02 H unit 
if care was exercised. 



3.2 

3.0 
2.6 
2.6 
2.4 
2.2 

> 2.0- 
55 18 
ui LG 

1.4 
1.2 
1.0 
0.8- 
06 
0.4 
Q2 




3.0: 
2.8- 
2.6 

2.4 
Z2 

2.0 
SL8 
| 14- 
1.4 
12 
LOh 
08 

0.4 
0.2 




3 



9 12 15 IS 21 24 27 30 

LOGE 
4 MINUTES 



3 6 9 It 16 It ft 24 21 30 

LOGE 
8 MINUTES 




3 



9 12 15 18 21 24 27 30 

LOG E 
Ifc MINUTES 



369 12 15 16 21 24 27 30 

LOG E 
32 MINUTES 



FIG. 5. Positive film in negative developer. 
Fresh developer 

O Aged developer 

A Synthetic aged developer 



(E) Iodide. The iodide test used was the one described in a pre- 
vious paper. 8 The importance of an analysis for iodide is shown in 
Fig. 3. The amount of iodide found in this aged developer was 3.5 
milligrams of potassium iodide per liter. 



202 EVANS, HANSON, AND GLASOE [j. s. M. P. E. 

(IV) MATCHING OF AGED AND FRESH DEVELOPERS 

Having then determined the errors in analysis which could be 
tolerated and having obtained methods of analysis which would give 



32 
3.0 

2. eh 

2.6 
2.4 
2.2 
2.0 



K 



0.8- 
0.6 - 
0.4 
0.2 



32 
30 
28 
26 
24 
2.2 
2.0 
,.8 
1.6 
1.4 
1.2 
1.0 
Q8 
0.6 
O.4 
0/2 




O 3 



fe 9 12 15 18 21 24 27 30 

LOG E 
I MINUTE 




03 6 9 12 15 18 21 24 27 30 

LOG E 
4 MINUTES 



32 
3.0 
2.8 

26 

24 

2.2 

2.0 

1.8 

>,.6 

Kl.4 

5 1.2 

iof- 

0.8 
06 
0.4 
02 




3 6 9 12 45 18 21 24 27 30 

L06 E 
2 MINUTES 




69 12 15 18 21 24 27 3O 

LOG E 
6 MINUTES 



FIG. 6. Positive film in positive developer. 
- - Fresh developer 
O Aged developer 
A Synthetic aged developer 



results within these limits, the next step was the matching of the 
sensitometric curves obtained with an aged and a fresh developer. 

(A) Aging of Negative Developer with Negative Film. Two hun- 
dred feet of negative film which had been flashed to give a developed 
density of about 1.0 were developed in 5 liters of fresh negative de- 
veloper for fifteen minutes. The developer then stood overnight and 



Feb., 1942] SYNTHETIC AGED DEVELOPERS 203 

was analyzed the next morning for elon, hydroquinone, bromide, and 
iodide. Then a fresh developer was made up using these analytical 
values. After about an hour, the pH of the aged developer was 
measured and acid added to the fresh developer to bring it to the 
same pH as the aged developer. Within a few hours exposed sensito- 
metric strips of negative film were developed in each of these de- 
velopers, using the technic described in Part I and the sensitometric 
curves compared (Fig. 4).* A third developer aged with negative 
film was used to develop sensitometric strips of positive film and com- 
pared to a fresh one made up to the analysis of the aged developer 
(Fig. 5). 

(B) Aging of Positive Developer with Positive Film. The same 
procedure given above was used in aging a positive developer with 
positive film which had been given a uniform exposure to give a de- 
veloped density of about 1.5. In order to build up the bromide con- 
centration to a value approximating that found in actual practice, 
some of the film was developed to completion in the light, otherwise 
very large quantities of film would be required. Three samples of 
positive developer which contained no potassium bromide were used 
and enough film was run through them to give bromide concentra- 
tions of 1.52, 2.28, and 3.80 grams per liter. Analyses for hydro- 
quinone, elon, and potassium iodide were performed, fresh developer 
was made up, the pH of fresh and aged developers adjusted to the 
same value, and exposed sensitometric strips of positive film were de- 
veloped in the two developers. The results obtained with the sample 
containing 2.28 grams of potassium bromide per liter are given in 
Fig. 6. 

SUMMARY 

This work, as far as it goes, indicates that in a used photographic 
developer there are no unknown compounds which affect its photo- 
graphic properties. An aged developer simply contains less of the de- 
veloping agents than before, and iodide and bromide have been added 
with a possible change in pH. 

This fact is of vital importance to control work since it is now pos- 
sible to state that an accurate analysis specifies the photographic 

* Comparisons made between developers which were used within a few hours 
and some which were allowed to stand overnight in stoppered bottles showed no 
difference. 



204 EVANS, HANSON, AND GLASOE [j. S. M. P. E. 

properties of the solution for a given use. Perhaps a word of caution 
should be added, however. The result obtained when an exposure on 
a photographic material is developed depends not only on the de- 
veloper formula of the film and the exposure but also, to just about 
as great an extent, on the physical conditions under which the de- 
velopment takes place. While chemical analysis can and will show 
whether or not a developer will give the same results as another de- 
veloper when used on the same equipment, it does not follow that if 
two developers are identical chemically they will give matched re- 
sults on two different developing machines. Temperature, agitation, 
aeration, fogging agents, direction of circulation and rate of drying 
all are capable of making just as great differences as a 5 or 10 per cent 
change in the elon concentration. For the control man whose job it 
is to maintain a particular developing machine constant, however, 
it has been demonstrated that if the elon, hydroquinone, bromide, 
iodide, and pH are correct the developer is chemically under control 
as far as the normal chemicals which affect the photographic results 
are concerned. Sudden onset of fog such as that due to sulfide bac- 
teria and the like, of course, are not checked. However, knowledge 
that all normal chemicals are correct speeds the discovery of the actual 
source of trouble. 

Another aspect of these results should be noted. It has been cus- 
tomary in the past to give the formulas used in various laboratories as 
though the fresh mixes were the important ones. It has long been 
known, of course, that they had little relation to the aged ones and 
would give different results if they were placed in the machines. 
Knowing now what the essential differences are, it becomes a simple 
matter to state accurately by analysis just what the actual developer 
is. A comparison of these actual formulas will eventually lead to 
quantitative technics of measuring the physical-photographic dif- 
ferences between different types of machines. 

For the practical laboratory man interested in maintaining his 
machine constant at all times and under all circumstances, this work 
suggests a valuable possibility. It is customary practice in most labo- 
ratories to shut down the machine occasionally for cleaning and for 
dumping the solutions, or at least part of them. Starting up the 
machine then has involved aging down the fresh solutions to the 
point where they matched previous results and getting the bath back 
to the control point. This is time-consuming and difficult, so much so, 
in fact, that many laboratories find it quite out of the question and 



Feb., 1942] SYNTHETIC AGED DEVELOPERS 205 

run for years with continuous replenishment without ever dumping 
the solutions. Both situations are bad and can now be avoided. If 
the actual formulas in use are known by analysis, a fresh mix which 
analyzes the same when it is placed in the machine, will give the same 
results without aging or any other special handling. Note, however, 
that it is the concentration of the compounds after the solution 
reaches the machine which must be the same as the standard. A 
great deal of elon and hydroquinone can be lost by oxidation while a 
machine is being filled. Note, also, that iodide as well as bromide 
must be added to such a mix. The same replenisher formulas should 
be made for such a mix as were used on the old batch. 

Acknowledgment. The thanks of the authors is due to Miss Jean- 
nette Klute and Mr. John Murphy for their assistance. 

DEVELOPER FORMULAS 

Negative Developer 

Fresh Aged 

I II 

G/L G/L G/L 

Sodium sulfite 100 100 100 

Elon 2.0 1.45 1.47 

Hydroquinone 5.0 3.7 4.40 

Borax 2.0 2.0 2.0 
Potassium bromide 1 . 07 . 98 
Potassium iodide 3 . 5 mg/1 1 . 8 mg/1 

(I) Used to develop negative film. 

(II) Used to develop positive film. 

Positive Developer 

Fresh Aged 

G/L G/L 

Sodium sulfite 45 . 45 . 

Elon 1.0 0.92 

Hydroquinone 4.0 2 . 75 

Sodium carbonate 30 . 30 . 

Potassium bromide 1.0 2 . 28 * 

Potassium iodide 0.4 mg/1 

*Developer contained no bromide before aging. 

REFERENCES 

1 JONES, L. A., RUSSELL, M. E., AND BEACHAM, H. R. : "A Developing Machine 
for Sensitometric Work," J. Soc. Mot. Pict. Eng., XXVIII (Jan., 1937), p. 73. 



206 EVANS, HANSON, AND GLASOE 

* EVANS, R. M., AND HANSON, W. T., JR.: "Chemical Analysis of an MQ 
Developer," /. Soc. Mot. Pict. Eng., XXXH (March, 1939), p. 307. 

3 PINNOW, J.: Zeitschr. wiss. Phot. (1912), p. 289. 

4 LEHMANN, E., AND TAUSCH, E.: Phot. Korr., 71 (Feb., 1935), p. 17; 71 
(March, 1935), p. 35. 

6 KOLTHOFF, I. M., AND LiNGANE, J. J. : Chem. Rev. 24, 1 (1939), p. 1. 

6 MULLER, OTTO H.: Chem. Rev., 24, 1 (1939), p. 95. 

7 BAUMBACH, H. L. : "The Chemical Analysis of Hydroquinone, Metol, and 
Bromide in a Photographic Developer," J. Soc. Mot. Pict. Eng., XXXIII (Nov., 
1939), p. 517. 

8 EVANS, R. M., HANSON, W. T., JR., AND GLASOE, P. K.: "Iodide Analysis 
in an MQ Developer," /. Soc. Mot. Pict. Eng., XXXVIII (Feb., 1942), p. 180. 






CURRENT LITERATURE OF INTEREST TO THE MOTION PICTURE 

ENGINEER 



The editors present for convenient reference a list of articles dealing with subjects 
cognate to motion picture engineering published in a number of selected journals. 
Photostatic copies may be obtained from the Library of Congress, Washington, D. C., 
or from the New York Public Library, New York, N. Y. Micro copies of articles 
in magazines that are available may be obtained from the Bibliofilm Service, Depart- 
ment of Agriculture, Washington, D. C., at prevailing rates. 



American Cinematographer 

22 (December, 1941), No. 12 

Using Arcs for Lighting Monochrome (pp. 558-559) 
Why Overlook the Set-Miniature? (pp. 560-561, 588) 
"Puppetoons" George Pal's Three- Dimensional Ani- 
mations (pp. 563, 588) 
Making Wipes in the Printer (pp. 574-575, 593) 

23 (January, 1942), No. 1 
Technical Progress in 1941 (pp. 6-7, 45-46) 
Increasing Focal Depth with the IR System (pp. 8-9, 

38-44) 

Coordinating Exposure-Meter and Processing for Effect- 
Lightings (pp. 10, 38) 

The Ah- Corps' Newest Camera Gun (pp. 11, 37-38) 

Determining the Sun's Angle for Any Location (pp. 
12-13, 37) 

Lamps Without Priorities (pp. 22-23, 34, 36) 

British Kinematograph Society Journal 

4 (October, 1941), No. 4 

Demonstrations of New Apparatus (pp. 143-149) 
Further Notes on a Kodelon Developer (pp. 150-154) 

Electricity in the Film Studio (pp. 155-166) 

Educational Screen 

20 (December, 1941), No. 10 

Motion Pictures Not for Theaters (pp. 427-^29), 
Pt. 32 

Electronic Engineering 

14 (December, 1941), No. 166 

Colour in Sound, a Study from the Psychological View- 
point (pp. 547-548) 



G. TOLAND 
V. KORDA 

A. WYCKOFF 
H. ZECH 



A. N. GOLDSMITH 

J. J. MESCALL 
R. N. HAYTHORNE 

L. KUTER 
W. STULL 



I. D. WRATTEN AND 

G. LEVENSON 
F. V. HAUSER 



A. E. KROWS 



E. L. TRIST 



207 



208 



CURRENT LITERATURE 



International Photographer 

13 (January, 1942), No. 12 
Outdoor Color Studio (p. 18) 

International Projectionist 

16 (October, 1941), No. 10 
Multiple-Speaker Reproducing Systems for Sound 

Motion Pictures (pp. 9-12, 14) 
Warner's "Vitasound" Theater System (pp. 14-16) 

Enlarging 16-Mm Kodachrome to 35-Mm Technicolor 

(P. 17) 
Magnetic Recording and Reproduction (pp. 18, 20) 

Motion Picture Herald (Better Theaters Section) 

146 (January 10, 1942), No. 2 
Getting Efficient Cooling in Spite of Priorities (pp. 

16-17, 24) 

Making the Theater a "Safe" Place in Wartime (pp. 
18-20) 



H. I. REISKIND 
N. LEVINSON AND 
L. T. GOLDSMITH 

W. STULL 



C. F. BOESTER 



FIFTY-FIRST SEMI-ANNUAL CONVENTION 

OF THE 
SOCIETY OF MOTION PICTURE ENGINEERS 



HOLLYWOOD-ROOSEVELT HOTEL 

HOLLYWOOD, CALIF. 
MAY 4th-8th, INCLUSIVE 



OFFICERS AND COMMITTEES IN CHARGE 

EMERY HUSE, President 

E. A. WILLIFORD, Past-President 

H. GRIFFIN, Executive Vice-President 

W. C. KUNZMANN, Convention Vice-President 

A. C. DOWNES, Editorial Vice-President 

J. G. FRAYNE, Chairman, Pacific Coast Section 

C. W. HANDLEY, Chairman, Local Arrangements Committee 

S. HARRIS, Chairman, Papers Committee 

Pacific Coast Papers Committee 

R. R. SCOVILLE, Chairman 

G. A. CHAMBERS F. L. EICH S. P. SOLOW 

C. R. DAILY W. W. LINDSAY, JR. W. V. WOLFE 

Reception and Local Arrangements 



J. O. AALBERG 
B. B. BROWN 
G. A. CHAMBERS 
W. E. GARITY 

A. M. GUNDELFINGER 

E. H. HANSEN 

J. K. HlLLIARD 

E. M. HONAN 



F. ALBIN 
L. W. CHASE 



C. W. HANDLEY, Chairman 

B. KREUZER 

R. G. LlNDERMAN 

C. L. LOOTENS 

R. H. McCULLOUGH 

W. C. MILLER 
G. S. MITCHELL 
K. F. MORGAN 
H. MOYSE 

Registration and Information 

W. C. KUNZMANN, Chairman 
J. FRANK, JR. 
J. G. FRAYNE 
C. W. HANDLEY 



W. A. MUELLER 
G. F. RACKETT 
H. W. REMERSHIED 
ALSTON RODGERS 
L. L. RYDER 
S. P. SOLOW 
H. G. TASKER 
J. R. WILKINSON 



S. HARRIS 
F. L. HOPPER 

209 



210 



L. A. AlCHOLTZ 

J. W. BOYLE 
J. L. COURCIER 



1942 SPRING CONVENTION 

Publicity 

JULIUS HABER, Chairman 
G. R. GIROUX, West Coast Chairman 
S. HARRIS 
S. E. HAWKINS 



[J. S. M. P. E. 



G. S. MITCHELL 
E. C. RICHARDSON 
R. R. SCOVILLE 



Luncheon and Banquet Committee 

L. L. RYDER, Chairman 
J. O. AALBERG EMERY HUSE 

J. G. FRAYNE H. T. KALMUS 

C. W. HANDLEY M. S. LESHING 

E. M. HONAN N. LEVINSON 



R. H. McCULLOUGH 

W. C. MILLER 

P. MOLE 

H. G. TASKER 



A. C. BLANEY 
D. J. BLOOMBERG 
L. F. BROWN 
J. P. CORCORAN 



J. O. AALBERG 

J. DURST 

G. M. FARLY 

B. FREERICKS 

W. E. GEBHART, JR. 



Hotel and Transportation 

G. A. CHAMBERS, Chairman 
C. R. DAILY 
C. DUNNING 
W. C. HARCUS 

G. T. LORANCE 

Convention Projection 

C. L. RUSSELL, Chairman 
L. D. GRIGNON 
J. K. HILLIARD 
A. E. JACKSON 
W. W. LINDSAY, JR. 

R. H. McCULLOUGH 



H. R. LUBCKE 

F. O'GRADY 

J. W. STAFFORD 
W. L. THAYER 



S. M. PARISEAU 
H. W. REMERSHIED 
C. R. SAWYER 
G. E. SAWYER 
H. A. STARKE 



Officers and Members of Los Angeles Projectionists Local No. 150 



Ladies' Reception Committee 

MRS. EMERY HUSE and MRS. J. G. FRAYNE, Hostesses 



Assisted by 



MRS. G. A. CHAMBERS 
MRS. F. L. EICH 

MRS. A. M. GUNDELFINGER 

MRS. C. W. HANDLEY 
MRS. J. K. HILLIARD 
MRS. E. M. HONAN 
MRS. B. KREUZER 
MRS. N. LEVINSON 
MRS. R. H. MCCULLOUGH 
MRS. G. S. MITCHELL 



MRS. 
MRS. 
MRS. 
MRS. 
MRS. 
MRS. 
MRS. 
MRS. 
MRS. 
MRS. 



P. MOLE 

K. F. MORGAN 

W. A. MUELLER 

G. F. RACKETT 

H. W. REMERSHIED 

E. C. RICHARDSON 

L. L. RYDER 

R. R. SCOVILLE 

S. P. SOLOW 

J. R. WILKINSON 



MRS. W. V. WOLFE 



Feb., 1942] 1942 SPRING CONVENTION 211 

Color Print Exhibit Committee 

O. O. CECCARINI, Chairman 

L. E. CLARK C. DUNNING L. D. GRIGNON 

J. B. CUNNINGHAM R. M. EVANS A. M. GUNDELFINGER 

TENTATIVE PROGRAM 

MONDAY, MAY 4, 1942 

9:00 a.m. Hotel Lobby; Registration 
12:30 p.m. Terrace Room; Informal Get-Together Luncheon 

Addresses by prominent Hollywood members of the motion picture 
industry; names to be announced later 
2:00 Blossom Room; General Session 

8:00 Blossom Room; Technical Session 

TUESDAY, MAY 5, 1942 

9:30 a.m. Hotel Lobby; Registration 

This morning will be devoted to a studio visit or other activity to 
be announced later 

2:00 p.m. Blossom Room; Technical Session 
8:00 Blossom Room; Technical Session 

WEDNESDAY, MAY 6, 1942 

9:30 a.m. Hotel Lobby; Registration 
10:00 Blossom Room; Technical Session 

2:00 p.m. The afternoon will be left open for a possible trip, to be announced 

later, or for recreation 
8:30 Blossom Room; Fifty-First Semi- Annual Banquet and Dance; details 

to be announced later 

THURSDAY, MAY 7, 1942 

10:30 a.m. Hotel Lobby; Registration 

Open morning 

2:00 p.m. Blossom Room; Technical Session 
8:00 Blossom Room; Technical Session 

FRIDAY, MAY 8, 1942 

10:00 a.m. Blossom Room; Technical Session 
2:00 p.m. Blossom Room; Technical Session 
8:00 Blossom Room; Technical Session 

Adjournment of the Convention 



212 1942 SPRING CONVENTION 

HEADQUARTERS 

The Convention headquarters will be at the Hollywood-Roosevelt Hotel. 
Excellent accommodations have been assured by the hotel management at the 
following per diem rates: 

One person, room and bath $3 . 50 

Two persons, double bed and bath 5.00 

Two persons, twin beds and bath 6.00 

Parlor suite and bath, one person 8 . 00-14 . 00 

Parlor suite and bath, two persons 12 . 00-16 . 00 

Room reservation cards will be mailed to the membership early in April and 
should be returned to the hotel immediately to be assured of satisfactory ac- 
commodations. 

Indoor and outdoor parking facilities adjacent to the hotel will be available for 
those who motor to the Convention. 

Golfing privileges may be arranged by request of the hotel management or at 
the registration headquarters. 

Registration headquarters will be in the hotel lobby. All members and guests 
attending the Convention will be expected to register and receive their Conven- 
tion badges. The registration fees are used to defray the expenses of the Con- 
vention, and cooperation in this respect is requested. Identification cards will 
be supplied, which will serve as admittance to all scheduled or special sessions, 
studio visits, and trips, and several de luxe motion picture theaters on Hollywood 
Boulevard in the vicinity of the hotel. 

Members planning to attend the Convention should consult their local railroad 
passenger agents regarding train schedules, rates, and stop-over privileges en 
route. For a stop-over at San Francisco the Convention Committee recommends 
the Mark Hopkins Hotel, on "Nob Hill." Accommodations may be arranged 
with Mr. George L. Smith, manager of this hotel. 

An interesting color-print exhibit will be an adjunct to the Convention and will 
be open to the public and delegates during the five days of the Convention. 

The Convention hostesses promise an interesting program of entertainment for 
the visiting ladies. A reception parlor will be provided as their headquarters at 
the hotel. 

Note: The Pacific Coast Section officers and managers gave serious considera- 
tion to the question of holding the 1942 Spring Convention at Hollywood, and 
have decided to proceed with arrangements for the meeting. The motion picture 
industry plays an essential part from the exhibiting and engineering viewpoint in 
upholding the morale of the general and theater-going public in the present crisis, 
and accordingly the Convention and Local Arrangements Committees are pro- 
ceeding with their plans. However, if later deemed advisable in the National 
interest, the Convention will be subject to cancellation thirty days prior to the 
announced Convention dates. 

W. C. KUNZMANN, 

Convention Vice-President 



SOCIETY ANNOUNCEMENTS 



FIFTY-FIRST SEMI-ANNUAL CONVENTION 

HOLLYWOOD-ROOSEVELT HOTEL 

HOLLYWOOD, CALIF. 

MAY 4-8, 1942 



The Board of Governors, the Board of Managers of the Pacific Coast Section, 
and the Convention Arrangement Committee have decided to hold the next 
Convention at Hollywood, May 4th to 8th, inclusive, as planned, subject to 
cancellation upon thirty days' notice if deemed necessary in the national interest. 
Committees of the Convention have begun to make their plans for the various 
sessions, and the Papers Committee has a number of interesting symposiums and 
presentations under consideration. Details concerning the sessions and the 
various Committees will be found on p. 209 of this issue of the JOURNAL. 



ADMISSIONS COMMITTEE 



At a recent meeting of the Admissions Committee, the following applicants for 
membership were admitted into the Society in the Associate grade: 



BADMAIEFF, ALEXIS 
c/o RCA Mfg. Co., Inc., 
501 N. LaSalle St., 

Indianapolis, Ind. 
BRECHA, H. C. 

348 Maryland Ave., 

Dayton, Ohio 
CLARK, R. 
65, Brampton Grove, 

Hendon, N.W.4, England 
DUVALL, D. P. 

419 W. College Ave., 

State College, Pa. 
HAMILTON, R. 

2921 Glenwood Ave., 
Minneapolis, Minn. 
HORNSTEIN, G. 
466 W. Fulton St., 
Long Beach, N. Y. 



KNOX, J. F. 
Bathurst Rd., 

Springwood, New South Wales, 

Australia 
MILLER, E. 
West End, 

North Carolina 
RAINBAULT, J. P. 
140 Riverside Drive, 
New York, N. Y. 
SEIDER, S. 

506 Avenue "S," 

Brooklyn, N. Y. 
WALTER, G. H. 
Quarry Heights, 

Canal Zone 
WHITE, D. K. 

223 Walton St.. N.W. 
Atlanta, Ga. 



213 



214 SOCIETY ANNOUNCEMENTS 

In addition, the following applicants have been admitted to the Active grade: 



ENGLER, R. J. 
41, Platt's Lane, 

London, N.W.3, England 



GRIFFITH, R. 

Museum of Modern Art Film Library, 
11 West 53rd St., 
New York, N. Y. 



BACK NUMBERS OF THE TRANSACTIONS AND JOURNALS 

Prior to January, 1930, the Transactions of the Society were published quar- 
terly. A limited number of these Transactions are still available and will be 
sold at the prices listed below. Those who wish to avail themselves of the op- 
portunity of acquiring these back numbers should do so quickly, as the supply 
will soon be exhausted, especially of the earlier numbers. It will be impossible 
to secure them later on as they will not be reprinted. 



1924 



1925 



No. 
19 

20 
21 
22 
23 
24 



Price 
$1.25 
1.25 
1.25 
1.25 
1.25 
1.25 



1926 



1927 



No. 

25 

26 

27 

28 

29 

32 



Price 
$1.25 
1.25 
1.25 
1.25 
1.25 
1.25 



1928 



1929 



No. 
33 
34 
35 
36 
37 
38 



Price 
$2.50 
2.50 
2.50 
2.50 
3.00 
3.00 



Beginning with the January, 1930, issue, the JOURNAL of the Society has been 
issued monthly, in two volumes per year, of six issues each. Back numbers of 
all issues are available at the price of $1.00 each, a complete yearly issue totalling 
$12.00. Single copies of the current issue may be obtained for $1.00 each. 
Orders for back numbers of Transactions and JOURNALS should be placed through 
the General Office of the Society and should be accompanied by check or money- 
order. 



SOCIETY SUPPLIES 

The following are available from the General Office of the Society, at the prices 
noted. Orders should be accompanied by remittances. 

Aims and Accomplishments. An index of the Transactions from October, 
1916, to December, 1929, containing summaries of all articles, and author and 
classified indexes. One dollar each. 

Journal Index. An index of the JOURNAL from January, 1930, to December, 
1935, containing author and classified indexes. One dollar each. 

Motion Picture Standards. Reprints of the American Standards and Recom- 
mended Practices as published in the March, 1941, issue of the JOURNAL; 50 cents 
each. 

Membership Certificates. Engrossed, for framing, containing member's name, 
grade of membership, and date of admission. One dollar each. 

Journal Binders. Black fabrikoid binders, lettered in gold, holding a year's 
issue of the JOURNAL. Two dollars each. Member's name and the volume 
number lettered in gold upon the backbone at an additional charge of fifty cents 
each. 

Test- Films. See advertisement in this issue of the JOURNAL. 



JOURNAL OF THE SOCIETY OF 
MOTION PICTURE ENGINEERS 

VOLUME XXXVIII MARCH, 1942 

CONTENTS 

The Color of Light on the Projection Screen 

M. R. NULL, W. W. LOZIER, AND D. B. JOY 219 

New 13.6-Mm Carbons for Increased Screen Light 

M. T. JONES, W. W. LOZIER AND D. B. JOY 229 

The Consumption of the Positive Arc Carbon 

H. G. MACPHERSON 235 

Stabilized Feedback Light- Valve 

W. J, ALBERSHEIM AND L. F. BROWN 240 

A Constant-Torque Friction Clutch for Film Take-Up 

W. HOTINE 256 

Recent Developments in Projection Mechanism 
Design E. BOECKING AND L. W. DAVEE 262 

Report of the Studio Lighting Committee 281 

Adventures of a Film Library R. GRIFFITH 284 

Work Simplification Essential to Defense 

A. H. MOGENSEN 295 

Current Literature 300 

Fifty-First Semi-Annual Convention, Hollywood, 
Calif., May 4-8, 1942 302 

(The Society is not responsible for statements of authors.) 



JOURNAL OF THE SOCIETY OF 
MOTION PICTURE ENGINEERS 

SYLVAN HARRIS, EDITOR 
Board of Editors 

ARTHUR C. DOWNES, Chairman 

JOHN I. CRABTREE ALFRED N. GOLDSMITH EDWARD W. KELLOGG 

CLYDE R. KEITH ALAN M. GUNDELFINGER CARLETON R. SAWYER 

ARTHUR C. HARDY 

Officers of the Society 

*President: EMERY HUSE, 

6706 Santa Monica Blvd., Hollywood, Calif. 
*Past-President: E. ALLAN WILLIFORD, 

30 E. 42nd St., New York, N. Y. 
" 'Executive Vice-President: HERBERT GRIFFIN, 

90 Gold St., New York, N. Y. 
^Engineering Vice-President: DONALD E. HYNDMAN, 

350 Madison Ave., New York. N. Y. 
^Editorial Vice-President: ARTHUR C. DOWNES, 

Box 6087, Cleveland, Ohio. 

* Financial Vice-President: ARTHURS. DICKINSON, 

28 W. 44th St., New York, N. Y. 

* Convention Vice-President: WILLIAM C. KUNZMANN, 

Box 6087, Cleveland, Ohio. 
^Secretary: PAUL J. LARSEN, 

1401 Sheridan St., N. W., Washington, D. C. 
* Treasurer: GEORGE FRIEDL, JR., 

90 Gold St., New York, N. Y. 

Governors 

*MAX C. BATSEL, 501 N. LaSalle St., Indianapolis, Ind. 
**FRANK E. CARLSON, Nela Park, Cleveland, Ohio. 

*JOHN G. FRAYNE, 6601 Romaine St., Hollywood, Calif. 

*ALFRED N. GOLDSMITH, 580 Fifth Ave., New York, N. Y. 
**EDWARD M. HONAN, 6601 Romaine St., Hollywood, Calif. 

*I. JACOBSEN, 177 N. State St., Chicago, 111. 
**JOHN A. MAURER, 117 E. 24th St., New York, N. Y. 

*LOREN L. RYDER, 5451 Marathon St., Hollywood, Calif. 

* Term expires December 31, 1942. 
** Term expires December 31, 1943. 



Subscription to non-members, $8.90 per annum; to members, $5.00 per annum, included 
in their annual membership dues; single copies, $1.00. A discount on subscription or single 
copies of 15 per cent is allowed to accredited agencies. Order from the Society of Motion 
Picture Engineers, Inc., 20th and Northampton Sts., Easton, Pa., or Hotel Pennsylvania, New 
York.N.Y? 

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 

General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 

Entered as second-class matter January 15, 1930, at the Post Office at Easton, 

Pa., under the Act of March 3, 1879. Copyrighted, 1942, by the Society of Motion 

Picture Engineers, Inc. 



THE COLOR OF LIGHT ON THE PROJECTION SCREEN* 
M. R. NULL, W. W. LOZIER, AND D. B. JOY** 

Summary. Spectral energy distribution measurements of the light at the center of 
the projection screen have been made for a number of arc lamp and carbon combina- 
tions employed in motion picture projection. ICI color-coordinates and color-tem- 
peratures have been calculated and show that the various high-intensity combinations 
give chromaticities and color -temperatures which agree closely among themselves. 
The spectral-energy distribution data of both low and high-intensity lamps and car- 
bons show marked similarity to black-body curves. 

The various combinations of carbon arcs and lamps employed for 
motion picture projection have been discussed before this Society 
on previous occasions. The characteristics of the arcs themselves 
and also their performance with complete optical systems have been 
considered and important factors concerning the light on the projec- 
tion screen have been determined. The growth of colored motion 
pictures has led to increased emphasis on the importance of color 
and we have undertaken a program of study of the color of the light 
on the projection screen in terms of modern methods of measurement 
and specification. 

A few years ago, members of the Research Laboratory of our Com- 
pany made measurements of the spectral-energy distribution of the 
light from carbon arcs used for photography in motion picture studios 
and also for projection in motion picture theaters. The results of this 
work have been presented by Bowditch and Downes. 1 ' 2 - 3 These 
measurements pertained to the direct radiation from the carbon arcs 
and in most cases referred to the crater radiation only. It was real- 
ized that in the case of carbon arcs used for projection, the passage 
of the light through the optical system would result in the selective 
absorption of light in certain wavelength regions and so alter the re- 
sultant color on the motion picture screen. Also the variations be- 
tween the color of the light emitted from the crater and the adjacent 

* Presented at the 1941 Fall Meeting at New York, N. Y. ; received No- 
vember 5, 1941. 

** National Carbon Company, Fostoria, Ohio. 

219 



220 M. R. NULL, W. W. LOZIER, AND D. B. JOY [J. S. M. P. E. 

portions of the arc are difficult to assess as regards their influence upon 
screen light. Since the factors described above are variables de- 
pendent upon the characteristics and adjustment of each particular 
optical system, their effect was not included in the earlier general de- 
scription of the radiation from the arc itself. In each installation, how- 
ever, the effect of the projector optical system on the color of the light 
must be recognized and we have recently evaluated the spectral- 
energy distribution of the light on the projection screen for a number 
of individual cases of lamps, optical systems, and carbons. These re- 
sults give us new and interesting information and, so far as we know, 
represent the first measurements of the spectral-energy distribution on 
the projection screen. 

The various projector arcs discussed below were burned in their 
respective lamps with the customary optical systems, illuminating a 
bare film aperture. A standard projection lens was used to focus an 
image of the film aperture on a miniature projection screen about one 
foot wide. The monochromator employed for measuring the spectral- 
energy distribution was placed with its entrance slit at the center of 
the illuminated projection screen. By means of the monochromator 
and associated thermopile and galvanometer we are able to deter- 
mine the relative amounts of energy in the various wavelength bands 
throughout the visible spectrum. While this method of measurement 
includes the effect of the lamp and projector optical systems upon the 
color of the light, it does neglect the influence of the projection screen 
and the motion picture film. If the projection screen is non-selective 
and reflects equal proportions of all wavelengths falling upon it, then 
it will not alter the form of the spectral-energy distribution curve or 
the color of the projection light. If the screen does have some spec- 
tral selectivity, the effect can be calculated by employing the spectral 
reflectance curve of the screen material in combination with 
spectral-energy distribution data reported in this paper. Similarly, 
the spectral transmission characteristics of the motion picture film 
can be combined with the data of this paper to determine the overall 
effect on the screen. The spectral-energy distribution data should 
therefore prove of particular value to those interested in the further 
development of colored motion pictures. 

Previous experience with carbon arc projector lamps has shown 
that when the arc is maintained at the correct distance from the re- 
flector or condenser lens the color over the projection screen has a 
uniform visual appearance. When the carbons are displaced from the 



Mar., 1942] COLOR OF LIGHT ON PROJECTION SCREEN 



221 



position of correct adjustment, there result changes in the intensity 
and distribution of the light on the screen which have been discussed 
in earlier publications. 4 ' 5> 6 If these displacements become severe 
enough they may first produce slight changes in color over the entire 
screen and, later, differences in color between different portions of the 
screen. It is hoped in the future to be able to present measurements 
on the extent of these color variations. However, for the measure- 
ments reported below, the carbon position has been maintained at 



/^u 
120 
100 
80 
60 

20 











x 


^x 

<T Low INI 

^T 30 / 


'ENSITY 
\MPS. 

AT 60AM 




jC g 




/;/" 


2 


^5: 


^.> 


PS. 


II 


/r* 


/ 


7 


\ /3.6H.l*r 

\ /25 AMPS.' 


^ 






/ 


7 


/ 


\ 


f^*'^Hu>n^ 


.ITY OF 
M FYF 




a| 






/ 










1! 
a ! 




2 


/ 




\ 








l/WLTL^ 


^W 




GREEN 


TS . 


<*M^ 

f RANGED 


K 




4000 5000 6000 7DOO 



WAVE LENGTH IN ANGSTROM UNITS 
FIG. 1. Spectral-energy distribution data for light at center of projec- 
tion screen with low-intensity lamp at 30 amps, Suprex-type lamp at 60 
amps, and condenser-type lamp at 125 amps, with heights of curves ad- 
justed to give equal foot-candles. 

the correct position to give uniform visual color over the screen and 
measurements have been carried out only at the center of the screen. 

The visible wavelength range extends from 4000 to 7000 A. In Fig. 
1 is shown the spectral-energy distribution over this visible range of 
light on the projection screen for three widely used projector lamps 
and arcs. These include the low-intensity lamp, the Suprex type 
lamp, and the condenser type lamp burning the 13.6-mm high-inten- 
sity carbon at 125 amperes. These spectral-energy distribution mea- 
surements tell nothing about the lesser amount of light obtained from 



222 



M. R. NULL, W. W. LOZIER, AND D. B. JOY [J. S. M. P. E. 



the low-intensity system ; in fact, for the purposes of this paper the 
heights of the three curves have been adjusted so they are on the 
basis of equal visible light. The spectral-energy distribution of the 
screen light from the low-intensity lamp shows the red radiation as 
the most plentiful, and relatively lesser amounts of green and blue. 
The two high-intensity lamps show on the screen a more even balance 
of energy among the different wavelengths with actually a slight pre- 
ponderance in the green region. 




30 . 40 50 60 70 80 
VALUES OF x 



-D.C i KW 

I NEW 136 M I 
-7 MM SUPSEX 




-8 MM SUP/IE.* 
-136 H. I. 



DOTTED SECTION 
MAGNIFIED 10 TIMES 



32 33 .34 35 36 37 38 39 
VALUES OF x 



FIG. 2. Chromatic! ty diagram showing ICI color-coordinates of various 
lamp and carbon combinations. The right-hand figure shows a tenfold 
enlargement of the indicated portion of the left-hand figure. 



These spectral-energy distribution data can be used to derive fur- 
ther quantities which are widely used in discussion and comparison 
of visual colors. Two such bases of comparison are (1) the chroma- 
ticity diagram, and (2) comparison with a black body. 

Chromaticity Diagram. The use of the ICI chromaticity diagram 
has been illustrated in a number of publications in recent years. 7 ' 8f 9 
According to this procedure, it is possible to calculate from the spec- 
tral-energy distribution of a light-source or illuminated object three 
numbers or color-coordinates which specify its color. 10 These are the 
so-called ICI trichromatic coefficients x, y, and 0. Since the sum of 
these trichromatic coefficients is unity, only two of them are necessary 
to describe the color. The coefficients x and y can be plotted as co- 
ordinates on a chromaticity diagram as shown in Fig. 2. The color- 



Mar., 1942] COLOR OF LlGHT ON PROJECTION SCREEN 223 

coordinates of all pure spectrum colors are known, and when plotted 
in this manner fall on the curved boundary of this diagram, with the 
wavelengths of the various parts of the spectrum indicated in A. All 
composite colors, which are in reality composed of varying proportions 
of pure spectrum radiation, will fall within the curved boundary of 
the chromaticity diagram of Fig. 2. 

The ICI color-coordinates of the various light-sources studied in our 
tests have been calculated and are given in Table I. Values for rep- 
resentative currents for each of the combinations shown in Table I 
have been plotted in Fig. 2. It is apparent, particularly from the en- 
largement shown at the right-hand side of Fig. 2, that the colors of the 
screen light from all the high-intensity arc lamps are in a closely 
bunched group which is distinctly separated from the color of the 
low-intensity arc lamp. This chromaticity diagram has the useful 
property that if the points representing any two component colors are 
connected by a straight line, then the points representing all possible 
combinations of these components will lie on that straight line. For 
example, on Fig. 2, a straight line drawn from the center of the 
group of "high- in tensity" points through the "low-intensity" point 
intersects the boundary of the diagram at 5840 A, in the yellow part of 
the spectrum. This explains the color differences observed when 
these two light-sources are-projected side by side. The ' 'low-intensity" 
color can be obtained by adding yellow light of wavelength 5840 A 
to the "high-intensity" light and therefore the "low-intensity" light 
appears yellow compared to the "high-intensity." 

Comparison with Black-Body Radiation. It is common procedure 
to compare the color of a so-called continuous-type light-source with 
that of a theoretical black body, the quality of whose radiation depends 
only upon the temperature. This comparison is expressed as the color- 
temperature of the light-source, which is defined as the temperature 
at which a black body would have to be maintained in order that its 
radiation would most nearly match the color of the source in question. 
Using this method of comparison, the color-temperatures of the screen 
light from the various combinations of lamps and carbons have been 
determined 11 as shown in Table I. We note that the color- tempera- 
ture of the screen light from the low-intensity lamp is 3870 K, while 
the high-intensity screen light ranges from 5020 to 5620K. The 
color- temperatures of the Suprex type, the "One Kilowatt" a-c and 
d-c, and the condenser- type lamps over their recommended current 
ranges all fall within this range. 



224 



. R. NULL, W. W. LOZIER, AND D. B. JOY [J. S. M. P. E. 



W 



o o o o 

CM 00 CD <N 

O '-i O CO 

1O lO *O O 



iOiO'OCOiOcO l 'O>O l O l OcO'OiOiOTHOO 

cococococococococococococococococo 
ooooooooooooooooo 



OOCO'-iCOOOT*COOQoqcOOOCq<N 
>OOCOOOO5O(NOOO5r^ 
| OOCOCOCOCOC v 'lcOCQCMCO T t l 

cocococococococococococococo 

OOOC5OCJOOOOCOOOCOCOOCJ 



^fTt(iO>OcOCDCOt^ 



o o 10 

CO CM ^ 






.1 



a 
d 

0} 



ex 

I 



u 

I 

ft 

- 



a 



ffi 



g 



o o 

o o 



HH 

^3 43 

.SP .SP 

W W 







2- 



s a 







a a a 



St^co 



i ' 5 






'! 



W 



8 .2 

1 = 

I! 



All lam 
Values 



Mar., 1942] COLOR OF LIGHT ON PROJECTION SCREEN 



225 



The specifications of light-sources by chromaticity or color-tem- 
perature are visual ones, based upon eye response and not upon the 
spectral-energy distribution alone. Thus two sources may match per- 
fectly in chromaticity or color-temperature, even though the spectral- 
energy distributions of their radiant energies are widely different. 



JJ 
20 
10 

20 
10 

20 
JO 



V3 
M 








P^ 

c/f 3opy 
3870 Y 


| 




(-LOW \ 

AT 3C 


NTENSITY 

AMPS. 


2 

tu 

U4 

K 




^^ 


AT 




i 
















e 




--~Z^~~ 


^ 


' -*. 


_.. ^ 






> 


^ 


^B 

AJ 


L>4Cr ^OPV 

5360 V 


8 MM 
AT 6C 


. 3UPJ?EX" 
) /4MPS. 


+^~~^~- ^ 




sJ 
tu 
















S 
tot 


_ 


--^ 


c/c 3opr 
>6/0 ')( 


^ * 

13.6 HI. 
^T 125 A 


^^^^T"~ 






I 


7 


<L R| >3 

AT ' 


MP5 


~~ ' 




ul 
(V 


VIOLET 


J*LUE 


Yt 
GREEN 


LLOW-) 

llM 


^MW R 


FP 




4000 5000 6000 7000 
WAVE LENGTH iw /AWGSTPOM I/NITJ 



FIG. 3. Comparison of spectral-energy curves of Fig. 1 with black-body 
curves of nearest chromaticity, with ordinates adjusted to give equal foot- 
candles. 

Therefore the practical significance of chromaticity and color-tem- 
perature is limited. For a black body, of course, the temperature com- 
pletely specifies the energy distribution as well as the color. It is 
interesting to note that on account of the close correspondence of 
the carbon arc light-sources under consideration to black-body 
sources, the color-temperature is an unusually good measure of the 
spectral-energy distribution in these cases as well. Fig. 3 shows 



226 M. R. NULL, W. W. LOZIER, AND D. B. JOY [J. S. M. P. E. 

the spectral-energy distribution of the "low-intensity," "Suprex," 
and condenser-type high-intensity screen light in comparison with 
spectral-energy distribution curves for black bodies at the same 
color-temperature and same candle-power. The spectral-energy dis- 
tribution of the screen light from the low-intensity projector lamp 
corresponds quite closely through the visible to that of a black body 
at the same color-temperature. With the high-intensity lamps the 
similarity is still almost as good though these arcs have slightly less 
energy in the red and blue and more in the green than the correspond- 
ing black bodies. 

In Table I are shown also color-temperature values obtained in 
the earlier measurements on the direct crater radiation from some of 
these arcs. 3 This has been unaltered by transmission through any 
optical system and comparison of these values with our values for the 
color-temperature of the screen light gives interesting information 
on the changes in color produced by the optical systems of the par- 
ticular projector lamps used. With the low-intensity lamp and the 
condenser-type high-intensity lamp the color-temperatures of the 
light on the screen show relatively small departure from that ob- 
tained on the bare sources. The low-intensity shows a little higher 
color-temperature for the light on the screen compared to the bare 
source, and the condenser-type high-intensity lamp shows slightly 
lower color-temperatures on the screen. With the 7-mm and 8-mm 
Suprex positive carbons the color-temperature of the bare source 
ranged from 5800 to 6400K. These light-sources show a color-tem- 
perature through the optical system on the screen 800 to 900 lower 
than the bare sources. This behavior brings the color-temperature 
of the light on the screen and its spectral-energy distribution into 
closer agreement with that obtained from the condenser-type lamp 
and results in a narrow spread of color between all our high-intensity 
arcs. 

There are other factors which can affect the spectral-energy dis- 
tribution and color- temperature of the light on the projection screen. 
We have made some preliminary studies of the effect of different 
lenses and mirrors, and believe that their effect in general will be 
small. As mentioned above, the position of the carbon with respect 
to the optical system can influence the color. Increase of carbon dis- 
tance from the mirror in general causes an increase in color-tempera- 
ture, although from exploratory measurements this change was of the 
same order of magnitude as the differences between the various high- 



Mar., 1942] COLOR OF LIGHT ON PROJECTION SCREEN 227 

intensity combinations shown in Table I. While these minor varia- 
tions in color-temperature of the screen light from the various high- 
intensity projector lamp and carbon combinations could probably 
be discerned by simultaneous comparison side by side, the differences 
are small in comparison with the familiar difference in color of screen 
light from the low-intensity and high-intensity lamps. 

If subsequent developments, especially in the art and technology 
of colored motion pictures, indicate the desirability of alterations in 
the color-temperature of some of these high-intensity lamps, this can 
to some extent be carried out. Changes in color-temperature can be 
produced by alteration of the carbon, which has been done for some 
applications in the past. 12 - 13 

The results described in this paper give us assurance that with 
equipment in good condition and properly adjusted, the popular 
high-intensity lamp and carbon combinations give remarkably con- 
sistent color and spectral-energy distribution. We plan to extend 
these measurements to a study of the variations in color that may be 
produced when the lamps and carbons are not maintained in optimal 
adjustment. 

REFERENCES 

1 BOWDITCH, F. T., AND DOWNES, A. C. : "The Photographic Effectiveness 
of Carbon Arc Studio Light Sources," J. Soc. Mot. Pict. Eng., XXV (Nov., 1935), 
p. 375. 

2 BOWDITCH, F. T., AND DOWNES, A. C.: "The Radiant Energy Delivered 
on Motion Picture Sets from Carbon Arc Studio Light-Sources," /. Soc. Mot. 
Pict. Eng., XXV (Nov., 1935), p. 383. 

3 BOWDITCH, F. T., AND DOWNES, A. C.: "Spectral Distributions and Color- 
Temperatures of the Radiant Energy from Carbon Arcs Used in the Motion 
Picture Industry," /. Soc. Mot. Pict. Eng., XXX (April, 1938), p. 400. 

4 JOY, D. B., AND GEIB, E. R.: "The Relation of the High-Intensity A-C 
Arc to the Light on the Projection Screen," J. Soc. Mot. Pict. Eng., XXIII (July, 
1934), p. 35. 

5 JOY, D. B., AND GEIB, E. R.: "The Non-Rotating High-Intensity D-C Arc 
for Projection," /. Soc. Mot. Pict. Eng., XXIV (Jan., 1935), p. 47. 

6 JOY, D. B., LOZIER, W. W., AND SIMON, R. W. : "Large Size Non-Rotating 
High-Intensity Carbons and Their Application to Motion Picture Projection," 
/. Soc. Mot. Pict. Eng., XXXIV (Mar., 1940), p. 241. 

7 HARDY, A. C.: "Handbook of Colorimetry," Mass. Inst. of Tech. Press 
(Cambridge, Mass.), 1936. 

8 MACADAM, D. L.: "The Fundamentals of Color Measurement," /. Soc. 
Mot. Pict. Eng., XXXI (Oct., 1938), p. 343. 

9 MOON, PARRY: "Color Determination," Ilium. Eng., XXXVI (Mar., 1941), 
p. 313. 



228 M. R. NULL, W. W. LOZIER, D. B. JOY 

10 BOWDITCH, F. T., AND NULL, M. R. : "Selected Ordinates for Computing 
Trichromatic Coefficients and Candle-Power of a Light-Source," /. Opt. Soc. 
Amer., XXVIII (Dec., 1938), p. 500. 

11 JUDD, D. B.: "Estimation of Chromaticity Differences and Nearest Color- 
Temperature on the Standard 1931 ICI Colorimetric Coordinate System," 
/. Opt. Soc. Amer., XXVI (Nov., 1936), p. 421. 

12 JOY, D. B., LOZIER, W. W., AND ZAVESKY, R. J. : "Recent Improvements in 
Carbons for Motion Picture Studio Arc Lighting," /. Soc. Mot. Pict. Eng., 
XXXIII (Oct., 1939), p. 374. 

13 LOZIER, W. W., AND JOY, D. B.: "A Carbon Arc for the Projection of 
16-Mm Film," J. Soc. Mot. Pict. Eng., XXXIV (June, 1940), p. 575. 



NEW 13.6-MM CARBONS FOR INCREASED SCREEN LIGHT* 
M. T. JONES, W. W. LOZIER AND D. B. JOY** 



Summary. A new 13.6-mm super high-intensity carbon designed for 170- 
ampere operation gives a substantial increase in light over either the old 180-ampere 
13.6-mm super carbon or the 125 to 150-ampere new regular 13.6-mm carbon de- 
scribed previously. 6 The new regular and super carbons are compared with the old 
carbons as to light available on the screen and as to efficiency of light production. 

During the past several years there have been two types of 13.6- 
mm carbons available for use in the condenser type of lamp with a 
rotating positive carbon the so-called regular carbon ordinarily 
burned at 125 amperes, and the super carbon burned at 180 am- 
peres. 1 - 2 However, the larger motion picture screens could not be 
lighted to adequate brightness with these carbons and the available 
optical systems. Recent improvements in carbons and optical sys- 
tems have radically increased the obtainable screen light intensities. 
This paper describes a new carbon designed to yield a higher amount 
of screen light than it is possible to obtain from the -other standard 
projector carbons, and discusses its possibilities for the illumination 
of large screens. 

Employing f/2.2 condensers and an //2.0 projection lens having 
treated surfaces, 3 the 13.6-mm regular carbon at 125 amperes gave 
approximately 11,500 lumens on the screen with 80 per cent side-to- 
center distribution and with the shutter not running. With a 90- 
degree projector shutter and a screen 30 feet in width this would 
amount to about 10 foot-candles in the center of the screen. For a 
flat white screen in good condition this is equivalent to a brightness 
of about 7.5 foot-lamberts, a figure which is below the recommended 
limits 4 of 10 J. For these theaters there was available the 13.6-mm 
super carbon at 180 amperes which gave 30 per cent more light and 
would therefore increase the foot-candle reading to 13 and the bright- 
ness to 10 foot-lamberts at the center of the 30-ft screen. The fact 

* Presented at the 1941 Fall Meeting at New York, N. Y. ; received November 
20, 1941. 

** National Carbon Company, Fostoria, Ohio. 

229 



230 



M. T. JONES, W. W. LOZIER, AND D. B. JOY [j. s. M. P. E. 



that this super carbon burns at 25 inches per hour compared to 13 
for the regular, and requires 180 amperes instead of 125, limits its 
application although it has been adopted by a number of the larger 
theaters and also has been used successfully for rear projection in the 
motion picture studios. 5 It has been possible to improve this situa- 
tion, as described last Spring, 6 through the development of a carbon 
which gives slightly higher light at 150 amperes than the super at 180 
amperes but with a burning rate of only 15 inches per hour instead 
of 25. It can also be burned at as low a current as 125 amperes and 
at this current gives the same light as the above-mentioned regular 
carbon but at a 35 per cent lower consumption rate. This has there- 

TABLE I 

Characteristics of 13.6-Mm H. I. and Super H. I. Projector Carbons under Typical 

Operating Conditions 



Carbon 

Arc amperes 

Arc volts 

Positive consumption rate 

(inches per hour) 
Crater candlepower 
Screen lumens without 

film shutter* 



Old 
Regular 
H.I. 
Projector 


Old 
Super H.I. 
Projector 


New Regular* 
H.I. 
Projector 8 


New Super 
H.I. 
Projector 


125 
68 


180 
75 


150 

78 


170 
75 



13 
43,000 

11,500 



25 

60,000 

15,000 



15 
63,000 

16,000 



22 

78,000 

18,500 



* At 80 per cent side-to-center distribution ratio with //2.2 condenser system 
and 5-inch //2.0 treated Super Cinephor projection lens. 

fore brought a center brightness of 10 foot-lamberts on a 30-ft flat 
white screen within the reach of many theaters which, because of 
either the high-carbon consumption or the high-arc current, hesitated 
to adopt the 180-ampere super carbon. 

However, there is an appreciable number of large theaters which 
have screens of 30 feet or wider where the desire is not for a carbon 
which will give the same light at a lower current or lower consump- 
tion, but for a carbon which will give more light and still be accept- 
able in respect to current usage and carbon consumption. We have 
recently been able to develop such a carbon. This is known as the 
new super carbon; it can be burned up to 170 amperes, where it has 
a consumption of 22 inches per hour and, with the same optical sys- 
tem described above, it gives at least 20 per cent more light on the 
projection screen than the 180-ampere super carbon. This makes 



Mar., 1942] CARBONS FOR INCREASED SCREEN LlGHT 



231 



available 18,500 lumens without film or shutter, and results in ap- 
proximately 16 foot-candles or 12 foot-lamberts at the center of a 30- 
ft screen, thereby approaching more nearly to the illumination de- 
sired by these large theaters. With optical systems using slightly 




8 64 2 024 6 8 
RADIUS OF CRATER IN MILLIMETERS 

FIG. l(a). Intrinsic brilliancy distribution across crater of 13. 6 -mm H. I. 

carbons. 

slower condensers and objective lenses, the increase in screen light 
with this new super carbon is in some cases as much as 35 per cent 
over that of the old super carbon. 

The color of the light on the projection screen is the same with the 
new super carbon as with the old. The light passing through the 
aperture, however, has less heating effect per unit of light. Using 
the same optical system with which the new super carbon gave 20 



232 



M. T. JONES, W. W. LOZIER, AND D. B. JOY [j. s. M. P. E. 



per cent more light than the old, measurements indicate only about 
5 per cent increase in total energy passing through the film aperture. 
A comparison of carbon consumption rate, current, arc voltage, 
and screen light for the four above-mentioned carbons is summarized 
in Table I. The increase in light with the new carbon is due to the 
higher and broader intrinsic brilliancy curve as indicated in Fig. 
This intrinsic brilliancy curve shows the amount of light 



CANWXTOWER 

PER 

SQUARE 
MILLIMETER 




6420246 
RADIUS or CRATER IN MILLIMETERS 

FIG. 1(&). Diagram of crater opening of 13.6- 
mm old regular and new super H. I. carbons show- 
ing brilliancy at center and near edge of crater. 

emitted in the forward direction per unit area across the crater. 
For example, assume we are looking directly into the crater as shown 
in Fig. l(b). From each sq-mm (about the size of the head of a com- 
mon pin) of area at the center of the crater of the new super carbon, 
the amount of light coming toward us would be equivalent to that 
from 940 candles. At this same point on the old type regular car- 
bon we would have the lower intensity of 685 candles. Similarly, 
near the side of the crater the brilliancy would be 380 candles per 
sq-mm for the new super and 200 for the old regular carbon. For a 
given optical system, i. e., condensers and objective lens, the light 



Mar., 1942] CARBONS FOR INCREASED SCREEN LIGHT 



233 



on the screen, as shown by Cook, 7 should be governed by the in- 
trinsic brilliancy of the carbon plus, to some extent, the width of the 
high-brilliancy usable portion of the crater. It is thus apparent why 
the available lumens on the screen for a given optical system and 



140 


NEW 
REGULAR 


120 
100 


F 

A' 


OLD 


AT 150 AMPS. 


NEW 
SUPER 

J70AHP3. 


(EGULAR 
M25AWS. 




K 




OLD 


80 







SUPER 

AT 180 AMPS. 










60 


- 


















40 


- 








A 










20 










































OLD NEW ^ R 


120 


REGULAR REGULAR iT nruMP* 


100 


AT 


I25w 


IPS. OLD AT 
SUPER 


I50AT 


IPS. " 


1 1 w r 


1 II v* 












AT I80AHPS. 










80 




















60 




















40 










B 










20 




















n 





















FIG. 2. Relative quantity of screen light: (A) per inch 
of carbon; (B) per arc kilowatt-hour. 

screen distribution are highest for the new super, somewhat lower for 
the old super and new regular, and still lower for the old regular. 

Another way of comparing these carbons is by a previously de- 
scribed method 8 * 9 of indicating the quantity of screen light produced 
per unit of carbon and electrical energy consumed. Such a compari- 
son is given in Fig. 2. The old super carbon achieves higher screen 



234 M. T. JONES, W. W. LOZIER, AND D . B . JOY 

light than the old regular at the expense of a 32 per cent decrease in 
quantity of light per inch of carbon consumed and an 18 per cent de- 
crease in light per arc kilowatt-hour. The new regular carbon at 
150 amperes with slightly higher light than the old super at 180 am- 
peres gives approximately 75 per cent more light than the old super 
for each inch of carbon consumed and is, in fact, about 20 per cent 
better in this respect than the old regular. The new super with its 
20 per cent increase in light over the old super is 40 per cent superior 
to the old super in quantity of light produced per inch of carbon. 
Even with its higher light output, the new super produces more light 
per arc kilowatt-hour than do any of the other carbons. This new 
super carbon therefore supplies a desirable increase in screen light 
with improved efficiency of utilization of carbon and power. 

REFERENCES 

1 JOY, D. B., AND DOWNES, A. C.: "Characteristics of High-Intensity Arcs," 
J. Soc. Mot. Pict. Eng., XIV (March, 1930), p. 291. 

2 JOY, D. B.: "A 13.6-Mm Super High-Intensity Carbon for Projection," 
J. Soc. Mot. Pict. Eng., XXVII (Sept., 1936), p. 243. 

3 RAYTON, W. B.: "New Lenses for Projecting Motion Pictures," J. Soc. 
Mot. Pict. Eng., XXXV (July, 1940), p. 89. 

4 "Report of the Standards Committee," /. Soc. Mot. Pict. Eng., XXXVI 
(March, 1941), p. 266. 

6 JOY, D. B., LOZIER, W. W., AND NULL, M. R.: "Carbons for Transparency 
Process Projection in Motion Picture Studios," J. Soc. Mot. Pict. Eng., XXXIII 
(Oct., 1939), p. 353. 

6 JONES M. T., LOZIER, W. W., AND JOY, D. B.: "A New 13.6-Mm High- 
Intensity Projector Carbon," J. Soc. Mot. Pict. Eng., XXXVII (Nov., 1941), p. 539. 

7 COOK, A. A.: "A Review of Projector and Screen Characteristics and Their 
Effects upon Screen Brightness," /. Soc. Mot. Pict. Eng., XXVI (May, 1936), p. 
522. 

8 LOZIER, W. W., JOY, D. B., AND SIMON, R. W.: "A New Negative Carbon 
for Low-Amperage High-Intensity Trims," J. Soc. Mot. Pict. Eng., XXXV (Oct., 
1940), p. 349. 

9 LOZIER, W. W., CRANCH, G. E., AND JOY, D. B.: "Recent Developments in 
8-Mm Copper-Coated High-Intensity Positive Carbons," J. Soc. Mot. Pict. Eng., 
XXXVI (Feb., 1941), p. 198. 



THE CONSUMPTION OF THE POSITIVE ARC CARBON 
H. G. MacPHERSON** 



Summary. The consumption of the positive electrode of an arc between solid 
carbons in air results partly from evaporation and partly from oxidation. The oxida- 
tion is operative chiefly on the sides of the carbon, tapering the end of the electrode and 
thus producing a tip with a diameter 1 / 2 to 5 / 8 that of the original carbon. The con- 
sumption of the flat tip, or crater, however, is due almost entirely to evaporation. The 
evaporation rate is controlled by the pressure of carbon vapor at the crater surface and 
the mechanism of diffusion away from it. This diffusion was computed on the as- 
sumption that it is similar to that occurring in the evaporation from liquid drops of the 
same diameter as that of the carbon crater, and it was shown that the linear electrode 
consumption should be inversely proportional to the crater diameter. This was borne 
out approximately by measurements of consumption rates made near the overload cur- 
rent. The absolute values of the consumption rates are consistent with the hypothesis 
that the surface temperature of the positive crater is at or near the sublimation tempera- 
ture of carbon. 

In an arc between pure carbon electrodes the crater face of the posi- 
tive carbon has a remarkably uniform temperature when the arc 
current is at its maximum for peaceful operation of the arc. This 
temperature is independent of the size of the positive electrode and 
can be reproduced at will. The constancy of this temperature has 
led to the belief that the positive crater of the arc operates at the 
sublimation temperature of carbon. In an effort to gain some light 
on this question, an attempt has been made to calculate theoretically 
the consumption rate of the positive carbon and compare this calcula- 
tion with measurements. 

Fig. 1 is a picture of an arc between solid carbon electrodes. In 
this case the lower carbon is the positive electrode and is 7 mm in 
diameter. The upper electrode is 5 mm in diameter and the arc 
current is about 12 amperes. The negative carbon tapers to a 
rounded point and is consumed and shaped to this form largely by 
oxidation of carbon from its sides in combination with some evapora- 
tion from the tip. The positive carbon is tapered somewhat on the 

* Received February. 2, 1942. 
** National Carbon Company, Cleveland, Ohio. 

235 



236 



H. G. MACPHERSON 



[J. S. M. P. E. 



sides by oxidation, but there is a flat tip or crater that is consumed 
almost entirely by evaporation. 

The calculation of the consumption rate of the positive carbon was 
made on the theory that the rate of evaporation from the tip is con- 
trolled by the vapor pressure of carbon at the temperature of the 
crater and the laws of ordinary gaseous diffusion away from the crater. 
The diffusion was computed on the assumption that it is similar to 
that occurring in the evaporation from liquid drops of the same 

diameter as the carbon crater. 
The diffusion is then radial with 
spherical symmetry, and it was as- 
sumed that the diffusion takes place 
into one hemisphere only. The con- 
sumption of the positive carbon as 
calculated in this way is propor- 
tional to 

DMp 
RdT 




where D is the diffusion constant of 
the vapor at the arc temperature, 
M is the molecular weight of the 
diffusing gas, p is the vapor pressure 
of carbon at the crater surface, R 
is the radius of the crater tip, d is 
the density of the carbon in the 
electrode, and T is the effective 
temperature of the gas through 
which the diffusion takes place. 

Some difficulty was encountered 
in obtaining a value for the diffu- 
sion constant of carbon vapor used in this equation. The diffusion 
constants of various gases into air as given in the International Critical 
Tables were plotted as a function of molecular weight on logarithmic 
paper and it was found that a fairly good straight line was obtained 
with a slope indicating that the diffusion constant varies inversely as 
the square-root of the molecular weight of the gas, as is, of course, 
predicted by kinetic theory. Thus a diffusion constant could be 
picked off from this curve for carbon vapor at room temperature. 
According to kinetic theory, the diffusion constant should vary as the 
three-halves power of the temperature. Experimentally, however, 



FIG. 1. Arc between small solid 
carbons. 



Mar., 1942] 



POSITIVE ARC CARBON CONSUMPTION 



237 



measurements near room temperature have shown that D varies as a 
higher power of the temperature, up to T 2 . This greater variation of 
diffusion with temperature is associated with a change in the collision 
cross-section of the atoms and probably does not extend to higher 
temperatures. Therefore it was thought most satisfactory simply to 
multiply the diffusion constant for room temperature by a factor of 
1.35 and use the predicted kinetic theory variation of diffusion con- 
stant with temperature. This factor of 1.35 was obtained from a 



1.5 



1.0 



0.5 




2.5 T.O 

LOG DIAMETER - INCHES 



1.5 



FIG. 2. Carbon consumption, calculated and mea- 
sured. 



Sutherland equation for the dependence of the viscosity on the tem- 
perature, based on viscosity measurements on nitrogen up to 1000 K. 
Substituting the value thus obtained for the diffusion coefficient, 
the linear consumption of the carbon electrodes in inches per hour is 



C = 0.0038 



Rd 



The symbols have the same meaning as above, p is expressed in 
atmospheres, T is in K, R is in inches, and d is the density ingrams 
per cubic-centimeter. A numerical calculation of the consumption 
was made, assuming a molecular weight of 12, a temperature of 



238 H. G. MACPHERSON [j. s. M. P. E. 

3950 K, a density of 1.62 corresponding to graphite electrodes, and a 
vapor pressure of carbon of one atmosphere. This calculation is 
plotted as the diagonal straight line of Fig. 2. Also shown in Fig. 2 
are measurements made on a large number of arcs using positive car- 
bons with original diameters of Vis to 3 / 8 inch. Some of the points 
represent arcs in which the positive carbon was used as the lower elec- 
trode and others in which the positive was the upper electrode, and 
both amorphous carbon and graphite electrodes are represented. 
The arcs were run at a current just below overload, and the consump- 
tion and final crater diameter measured. The consumption rates 
varied from about 3 to 28 inches per hour in the range of carbons used. 
The measurements confirm the prediction that the consumption rate 
should vary inversely as the crater diameter. The actual consump- 
tions found are on the average about 30 per cent lower than those 
calculated. This must be considered a fortuitously close fit, con- 
sidering the approximate measurements and the uncertainty in the 
calculations. 

These uncertainties in the calculations should be pointed out. In 
the first place, it is assumed that the carbon burns with a hemispheri- 
cal tip instead of a flat tip. This assumption would lead to a high 
value for the calculated consumption. It was further assumed that 
the carbon vapor diffuses only throughout one hemisphere instead of 
in all directions. This probably tends to produce too low a value for 
the calculated consumption, although the carbon oxidized and evapo- 
rated from the tapered sides will prevent downward diffusion to some 
extent. Furthermore, the extrapolation of the diffusion constant to a 
new gas at an extremely high temperature is somewhat uncertain. 
The temperature of the gas in which the diffusion takes place is not 
uniform, but is higher than the assumed value within the arc stream 
and is lower than this value at some distance to the sides of the arc. 
In view of these approximations in the calculations, absolute agree- 
ment with experiment is not expected to be close, and therefore we 
can not hope to get an accurate value for the vapor pressure of carbon 
at the temperature of the crater surface from these data. However, 
the trend of the data indicates the correctness of our viewpoint that 
the loss of evaporated carbon is determined by a process of ordinary 
gaseous diffusion through a non-turbulent gas. Furthermore, the 
absolute agreement is such that it is probable that the vapor pressure 
of carbon at the temperature of the crater surface lies between four- 
tenths of an atmosphere and one atmosphere. 



Mar., 1942] POSITIVE ARC CARBON CONSUMPTION 239 

If we assume that, at the maximum crater temperature, the vapor 
pressure of carbon is one atmosphere, then a simple theory of over- 
loading of such an arc can be outlined. So long as the vapor pressure 
of carbon at the surface is equal to or less than atmospheric pressure, 
then a smooth streamline flow of gas around the positive carbon will 
be obtained and the carbon vapor will leave the crater surface by 
diffusion through this gas. However, if the current is raised to such a 
point that this sublimation temperature of carbon is exceeded at 
points on the crater, the excess pressure of carbon vapor developed 
will produce spurts of gas away from the crater face and cause a turbu- 
lence. The turbulent flow will allow fresh air with a low concentra- 
tion of carbon vapor to come close to the crater face and this will result 
in such a high concentrational gradient that the diffusion and evapora- 
tion of carbon from this part of the crater face will be extremely rapid. 
The rapid evaporation will, of course, cool the spot on the crater very 
rapidly, restoring that part of the crater to normal operation. At 
overload currents this process goes on repeatedly, resulting in an un- 
stable arc and in a sputtering or hissing noise. Just below this cur- 
rent, however, smooth operation is obtained, corresponding to stream- 
lined flow and steady diffusion of carbon vapor. 



STABILIZED FEEDBACK LIGHT-VALVE* 
W. J. ALBERSHEIM AND L. F. BROWN** 

Summary. Feedback affords controlled and undistorted damping of light-valve 
resonance. All electromechanical devices can be regarded as feedback circuits if their 
motional impedance is interpreted as a feedback counter-emf. Auxiliary amplifica- 
tion of this counter-emf produces stabilized motional feedback. Depending on 
whether the amplifier counter-emf is proportional to amplitude, velocity, or accelera- 
tion, the feedback tends to flatten amplitude, velocity, or acceleration response charac- 
teristic. The light-valve is a mechanically resonant device operated on an amplitude 
basis but with a velocity counter-emf. Velocity feedback increases the effective damp- 
ing of the light-valves; reactive components in the electrical driving impedance and in 
the feedback- gain tend to shift the resonance frequency. At 0.71 of critical damping 
the steady-state frequency characteristic is peakless and the valve follows transient 
impulses quickly with only 6 per cent overshooting. The maximum "bucking power" 
opposed to the valve motion by the feedback amplifier occurs at 0.58 of ribbon resonance 
and is 8 db less than the low-frequency power of the driving amplifier. 

An amplifier is described that has been designed in accordance with the theory for 
application of stabilized feedback to commercial light-valves. 

The feedback light-valve was developed in order to obtain con- 
trolled damping of mechanical resonance without the distortion and 
temperature variations inherent in mechanical damping methods. 

The application of stabilized feedback to light-valves is a special 
case of the well known negative feedback in transmission systems. 1 
Fig. 1 shows schematically two forms of electrical feedback. In Fig. 
l(a) the feedback-emf is obtained from the output voltage of the 
stabilized amplifier. It becomes inoperative if the output voltage is 
short-circuited by the switch indicated in the drawing. In Fig. 1 (V) 
the feedback-emf is obtained from the output current of the amplifier 
so that it becomes inoperative if the output current is interrupted by 
the switch SW. The gain of such feedback amplifiers is computed by 
equations 3 and 6 in the appendix to this paper. They are alike in 

* Presented at the 1941 Fall Meeting at New York, N. Y. ; ms. received Jan- 
uary 29, 1942. 

** Electrical Research Products Division of Western Electric Company, New 
York, N. Y., and Hollywood, Calif. 

240 



STABILIZED FEEDBACK LIGHT- VALVE 



241 



form although the physical meanings of the symbols ju and /3 in the 
two equations are different. 

In both cases, the polarity of the feedback is such that it reduces 
the amplifier gain. If the overall feedback gain n$ is much larger 
than one, the output tends to become independent of ju and is deter- 



SW\ OUT | LOAD 

RESISTANCE 




FIG. 1 (a). Feedback from output voltage. 

mined by the loss in the feedback network as shown by equations 
3b and 6b. Since amplifier gains vary with battery voltages, tube 
characteristics, and frequency, whereas the feedback circuit can be a 
passive network with a constant and predetermined loss character- 
istic, the amplifier is "stabilized" by reducing the influence of /*. 

Consider now an electromechanical device of arbitrary nature such 
as shown in Fig. 2. The input voltage E Q produces a motion of 




FIG. 1 (6). Feedback from output current. 

velocity V in the armature. Since the device is built for the sake of 
this mechanical motion, one can regard its velocity as the "output," 
and as corresponding to the output current of the electrical amplifier 
shown in Fig. 1 (6) . If the device is of an electromagnetic or electro- 
dynamic nature, the motion of the armature will produce in the elec- 
trical input circuit a counter-emf proportional to the armature 
velocity. This counter-emf disappears when the armature is clamped, 
just as the feedback-emf in Fig. l(b] disappears if the output current is 



242 



W. J. ALBERSHEIM AND L. F. BROWN [J. S. M. P. E. 



interrupted. Formally, the electromechanical device corresponds 
exactly to the current- type feedback amplifier of Fig. 1(&). In ac- 
cordance with this analogy, the mathematical formula for the output 
velocity (equation 12) shows the same structure as equation 6. The 
stabilizing effect of the electromechanical reaction is well known and 




CLAMP 



ARMATURE 



FIG. 2. Electromechanical device. 



utilized for the damping of electrical meters and other devices. It is 
usually, however, not interpreted as a feedback or counter-emf but as 
motional impedance. 

In order to enhance the effect of this reaction, one may pick up a 
part of the motional counter-emf, boost it in an auxiliary amplifier, 
and feed it back into the electrical input circuit. 2 This leaves the 
form of the equations unchanged but it increases the magnitude of the 





MAGNET 



L.V. RIBBON 



FIG. 3. Light-valve. 

feedback-constant so that one can obtain greatly increased stabiliza- 
tion. If the feedback input voltage is picked up electromagnetically 
or electrodynamically, it is proportional to the armature velocity and 
therefore tends to produce a constant- velocity characteristic. If the 
feedback pick-up is electrostatic, it depends upon the position of the 
armature and therefore tends to produce constant-amplitude charac- 
teristics. If the feedback pick-up generates voltage by pressure, it 
will tend to flatten the acceleration characteristic of the device. 



Mar., 1942] 



STABILIZED FEEDBACK LIGHT- VALVE 



243 



The general results given above are directly applicable to the ribbon 
light-valve used in sound-film recording. The light-valve is, in prin- 
ciple, a very simple device, as shown by Fig. 3. It consists merely of 
one or several tightly stretched metal ribbons placed in a magnetic 
field. Mechanically the light-valve ribbon is a resonant structure, 
with extremely small damping, so that if one eliminates the effect of 
its motional counter-emf by driving it at constant current (that is, 
infinite input impedance), its response shows a very sharp and high 
peak. The velocity-response for this condition is given by equation 
17. Since the use of the light- valve depends upon amplitude rather 
than velocity, its performance can be more readily analyzed from its 
amplitude characteristic, which is given by equation 18. If the valve 



500' 
IN 




FIG. 4. Schematic circuit of feedback light- valve. 



is operated from a finite electrical input impedance, its velocity and 
amplitude responses are given by equations 19 and 20, which differ 
from the previous equations by the addition of an electrodynamic 
damping term proportional to the square of the magnetic field 
strength and to the electrical conductivity. By very careful design 
it has been possible in recent light- valves to lower the amplitude peak 
from the 40 db corresponding to mechanical friction to less than 10 db. 
In most light-valves now in commercial use, however, the resonance 
peak approximates 20 db. Such a peak is undesirable, due not only 
to the steady-state frequency distortion but because it increases the 
danger of overload, and produces free vibrations at the ribbon's 
resonance-frequency when the valve is excited by a transient impulse. 
It was therefore decided some years ago to improve the light- valve 
characteristic by the application of stabilized feedback. The type of 



244 



W. J. ALBERSHEIM AND L. F. BROWN [J. S. M. P. E. 



circuit adopted is schematically shown in Fig. 4. 3 The problem of 
obtaining a pick-up voltage proportional to the ribbon motion, with- 
out changing the internal structure of the light-valve, was solved by 
means of a bridge circuit consisting of potentiometer P lt light-valve 
ribbon resistance r v , and the "simplex" resistance r s . This bridge is 
electrically balanced so long as the ribbons are prevented from moving 
(by clamping or by removal of the magnetic field). Under operating 
conditions, the voltage across the bridge is directly proportional to the 
ribbon velocity. It is boosted in the auxiliary amplifier and fed back 
into a low-impedance transformer winding in series with the main 
driving transformer. Since the voltage pick-up is proportional to 




N = FREQUENCY RELATIVE TO RIBBON RESONANCE 

FIG. 5. Light-valve velocity characteristic. 

velocity, one might expect that the feedback would produce a con- 
stant-velocity light-valve. This can actually be obtained, as shown 
by the velocity characteristic of Fig. 5, curve 4, but only at imprac- 
tically high feedback values and with great loss in efficiency over the 
entire working range. Fortunately a much smaller amount of feed- 
back is sufficient to produce a smooth amplitude characteristic at 
frequencies below the ribbon resonance. This is indicated by the 
amplitude characteristics of Fig. 6. By a moderate amount of feed- 
back, the damping factor can be made equal to 0.71 of the critical 
damping. At this damping the resonance peak is completely elimi- 
nated and transformed into a smooth low-pass filter characteristic 
which droops 3 db at the ribbon's resonance frequency, and at higher 
frequencies produces a loss of 12 db per octave (Fig. 6, curve 3). The 



Mar., 1942] 



STABILIZED FEEDBACK LIGHT- VALVE 



245 



transition from flat to drooping response is so gradual that it can easily 
be compensated by equalization, and the working range extended to 
frequencies well above light- valve resonance. 

Fig. 7 shows the transient responses corresponding to the steady- 
state velocity and amplitude characteristics, shown in Figs. 5 and 6. 
Curves 1 (Fig. 7) show the response of a light-valve damped only by 
mechanical friction. It is seen that a sudden impulse produces a 
weakly damped free oscillation which requires hundreds of vibrations 
before it dies down to imperceptible levels. Curves 2 correspond to a 
10-db amplitude peak, drawn on a much wider scale because the free 




FIG. 6. Light-valve amplitude characteristics. 

vibrations caused by a transient impulse die down after two or three 
cycles. Curve 3 corresponds to 0.71 of critical damping, which pro- 
duces a peakless, smooth, steady-state amplitude characteristic. The 
transient response overshoots its final position by only about 6 per 
cent and the transient dies down after a time, corresponding to one 
vibration of the undamped ribbon. Curve 4 is drawn for a feedback 
five times as large as critical damping. This produces not only a 
loss in efficiency and a drooping amplitude characteristic but also a 
very sluggish aperiodic response to transient impulses. 

The curves in Figs. 5, 6, and 7 are computed from equation 25, 26, 
and 31. These equations are strictly correct only if the electrical in- 
put to the light- valve is a pure resistance and the feedback amplifier 
gain is constant and accordingly free from phase distortion at all fre- 



246 



W. J. ALBERSHEIM AND L. F. BROWN [J. S. M. P. E. 



quencies. If the electrical input or the feedback-gain contain reactive 
components, the result will be a modification of the damping and a 
change in the resonance frequency of the light-valve. Under practi- 
cal conditions it is neither possible nor desirable to obtain completely 
flat feedback-gain characteristics. The feedback amplifier needs to be 
efficient only in a frequency-band near the ribbon resonance. At low 



J_4- 

DrO.OI 
yi|_ e -0.005x.cO5x 




20 40 



60 100 120 140 160 160 200 220 240 260 
CURVE -I 



3456749 10 II 12 13 




FIG. 7. Light-valve transients. 



frequencies its gain is greatly reduced in order to prevent feedback of 
noise-reduction bias currents; at high frequencies its gain droops to 
avoid singing at mechanical harmonics of the ribbon resonance fre- 
quency. In the working range its phase characteristic is such that it 
tends to raise rather than lower the resonance frequency of the light- 
valve and to produce an even flatter response than indicated by the 
elementary theory. 



Mar., 1942] 



STABILIZED FEEDBACK LIGHT- VALVE 



247 



The main effect of stabilized feedback from a steady-state point of 
view is a reduction of ribbon amplitude near resonance. The feed- 
back amplifier must therefore supply some electrical power opposing 
or "bucking" the power supplied by the driving amplifier. This 
bucking power is much smaller than the main driving power because 
at low frequencies the required input correction is small, and at reso- 
nance, where nearly the entire driving voltage must be bucked out, 
the current taken by the valve is negligible. As shown in the appen- 
dix, a maximum of 8 db less than the power of the driving power is 
required. Since this maximum occurs at high frequency (0.58 of 




FIG. 8. RA-1111-A amplifier. 



ribbon resonance), all harmonics of the feedback-emf fall beyond the 
transmission band of the reproducing circuits. The power require- 
ments of the feedback amplifier are therefore not severe. 

The ERPI RA-1111-A amplifier has been designed in accordance 
with the foregoing theory for the application of stabilized feedback to 
the RA-1061 light-valve as well as certain other ERPI light-valves. 
The schematic circuit of the amplifier is shown in Fig. 8. This device, 
of which Fig. 9 is a photograph, is contained in a dish-type, 5 l /t-in.. 
rack-mounting panel. The vacuum-tube heater and plate voltage are 
provided by a self-contained, regulated power unit which enables the 
amplifier to operate from 115-v, 50 to 60-cycle alternating current. 



248 W. J. ALBERSHEIM AND L. F. BROWN [J. S. M. P. E. 

The bridge-balancing potentiometer and electronic "eye" and am- 
plifier gain control are accessible through a door in the front mat. 
Balancing is accomplished by adjusting the electronic "eye" for 
minimum deflection with a 400-cycle sine- wave signal impressed on 
the input of the light-valve transformer. For this adjustment, the 
first two stages of the amplifier and the electronic "eye" act as the 
"galvanometer" in the bridge circuit. After this adjustment has 
been made, the gain of the amplifier is adjusted to give the same light- 
valve response or ribbon amplitude at 400 and 8000 cycles for equal 
levels at the light- valve transformer input. This may be done by a 




FIG. 9. RA-1111-A amplifier. 

stroboscopic method or by the use of a calibrated photoelectric cell 
monitor. The design of the amplifier circuit is such that this adjust- 
ment will result in uniform light- valve response from 40 to 8000 cycles 
for light-valves tuned to 10,000 cycles. The response decreases 
sharply above 10,000 cycles. 

Thus the use of a multisection low-pass filter is unnecessary with 
the stabilized feedback light-valve, and the transient distortions of the 
resonant circuits of the filter and the mechanical resonance of the 
light- valve are eliminated. The elimination of these distortions is 
easily demonstrated by the use of square-wave signals, and indeed 
square-wave signals are used in the production adjustment and testing 
of the amplifiers. The effect upon the ear of the removal of these 
distortions is not readily correlated with other distortion effects, and 



Mar., 1942] STABILIZED FEEDBACK LlGHT- VALVE 249 

has been described as resulting in a "more solid bass" and a "decrease 
in response in the 300-cycle region," which leads to the conclusion 
that the undesirable effects of the transient high-frequency distortions 
lie in the generation of intermodulation products by the high-fre- 
quency oscillations excited by impact signals in undamped circuits. 

The RA-1111-A amplifier may also be used to apply stabilized feed- 
back to push-pull light-valves. For this use the counter-emf from 
two ribbons of the valve is amplified and fed back to all four ribbons. 
The motional impedances of the various ribbons are sufficiently uni- 
form to make this practicable. 

REFERENCES 

1 BLACK, H. S. : "Wave Translation System" ; U. S. Patent 2,102,671. 

2 MAXFIELD, J. P., AND HARRISON, H. C. : "Vibratory System"; U. S. Patent 
1,535,538. 

3 ALBERSHEIM, W. J., AND SETTE, W. J.: "Recording System"; U. S. Patent 
2,157,880. 

APPENDIX 

(1) ELECTRICAL FEEDBACK CIRCUITS 

(1.1) Feedback from Output Voltage. 

Source voltage = e Q 

Feedback voltage = e/ 

Amplifier input voltage = e\ 
Amplifier output voltage e z 
Amplifier gain 



Feedback gain 



M 

in ( - ) 
W 



o / = /302 = e$ M/3^1 (l) 



eo 



1 + 



(2) 



for up 1, e, = l (3b) 



(1.2) Feedback from Output Current. 

Source voltage = eo 

Feedback voltage = e/ 

Amplifier input voltage = e\ 

Amplifier output current = iz 



Amplifier transconductancel - )= // 



250 W. J. ALBERSHEIM AND L. F. BROWN [J. S. M. P. E. 

Feedback impedance ( - ] = /3' 
W 

ei = e Q - ef = e<> P'i 2 = e n'&'ei (4) 



** = b 1 1 , R , ( 6 ) 

1 + AI ft 



for 



(2) ELECTROMECHANICAL DEVICES 

(-2..Z) Analogy of Electromechanical Device and Current Type Feedback Circuit. 

Mechanical force = F 

Velocity = v 

Mechanical impedance = Z m 

Electrical impedance = z 

Electrical input current = i 

Electrical input voltage = e 
Electromechanical counter-emf = e c 

Force factor = A 

F = Ai (7) 

i = Oo - e c )/z e (8) 

v = = T m (9} 

(e Q - e e )A 



(10) 
(11) 



V^O iwar/** "/ '-'*' A* / . n\ 




with 



ft' = A (14) 

(2.2) Application of Stabilized Feedback to Electromechanical Devices. Assume 
that by auxiliary amplification the dynamic counter-emf is boosted to 

e t = c e (l + b) = vA(l + b) (15) 



Mar., 1942] STABILIZED FEEDBACK LlGHT- VALVE 251 

Then, 

A_ 
[e - (1 + b)vA]A e Z m Ze r 

= 1 + (1 + JM s = N-T^< <") 

m 

in which /3'" is increased to 

/?'" = (1 + b)A (16b) 

(3) LIGHT-VALVE RESPONSE 

Electrical impedance of valve plus driving circuit = z e 

Magnetic field strength = B 

Effective valve ribbon length = L 

Effective flux BL (force factor) = A 

* Effective ribbon stiffness = S 

*Effective ribbon mechanical resistance = R 

*Effective ribbon mass = M 

(3.1} Constant- Current Input (Infinite Driving Impedance). 
Velocity: v = i~ = *-= ^ - (17) 



Amplitude: y = = ^ _ M + Rjw () 

This is sharply resonant at the frequency \/ ' M/S because the mechanical friction 
resistance is so small that a peak of 40 db may occur. 

(3.2) Voltage Type Input. With a finite driving impedance (preferably lower 
than the ribbon resistance) one finds immediately from 12: 

A A 2 

e - e 

(19) 






(20) 



S - Mw* + jwR + 
z 

(4) FEEDBACK LIGHT-VALVE (FIG. 4) 

Assume that the impedances of the bridging potentiometer P lf filter network F, 
and gain-control potentiometer P 2 are high compared to the light-valve and 
driving-circuit impedances. The bridge circuit is electrically balanced so that 
edyn is zero as long as the ribbons do not move (5 = 0). 

* Note: Effective stiffness, mechanical resistance, and mass are defined as the 
total force distributed over the effective ribbon length required to obtain at the 
center of the light-valve ribbon a unit of amplitude, velocity, and acceleration, 
respectively. 



252 W. J. ALBERSHEIM AND L. F. BROWN [j. S. M. P. E. 

When the ribbons move, ea y n is proportional to their velocity, and for a sym- 
metrical bridge (r = r v ) and very high impedance F or P 2 , it approaches half the 
dynamic counter-emf of the light-valve : 

*dyn = 0.5e 6 (21) 

The amplified counter-emf fed back into the driving circuit is 

e a = be e (22} 

where b is the product of network losses and amplifier gain. The total effective 
feedback voltage becomes 

e f = e e + e a = (1 + b)e c (15) 

Introducing this value into 12, 



1 + (1 + b) ~ 

*-*m. 



(16) 



-41 

JW ' " Z e 

A 

(24) 



S - Mw* + I R + (1 + b) 

If z e approximates a pure resistance r, and b a constant numerical gain without 
reactive components, equations 23 and 24 can be simplified and normalized : 

N 



(Fig - 6) 



D = I R + (I + b) 4- \(SM) ' (28) 

(29) 
(30) 

These steady-state velocities and amplitude response curves have been plotted 
in Figs. 5 and 6 for various values of the damping factor D. Curves 1 are drawn 
for D 0.01 which approximates a light-valve operated at constant current with 
a 40-db peak. Curves 2 have D = 0.316 corresponding to a 10-db peak. Curve 3 
is drawn for D = \/2, which just eliminates the amplitude peak. Its amplitude 



Mar., 1942] STABILIZED FEEDBACK LlGHT- VALVE 253 

response is flat up to one octave below resonance, has a 3-db loss at ribbon reso- 
nance, and droops 12 db per octave at high frequencies. Curve 4 shows the im- 
practical case of D = 10. Its response is inefficient and nearly flat on the velocity 
scale, drooping 6 db per octave on the amplitude scale. 

(5) TRANSIENT RESPONSE (FIG. 7) 

Related to the steady-state characteristic, per equation 26, is the response of the 
light- valve to "transients"; that is, shock impulses. The nature of these shocks 
may be electrical (switching clicks) or mechanical (overload clash). Assume that 
at t = the voltage jumps from zero to e . One may transform 26 into the opera- 
tional form 



y, 1+P* + DP 
with 

"''It (32} 

The abscissa unit x is such that a full vibration cycle of the undamped ribbons at 
their resonance frequency corresponds to an increase of x by 2?r. 

(5.1) Damped Oscillations. For D < 2 the transients are free damped vibra- 
tions according to the equation 

^L = 1 - ^.f-Dx/z . shi (F X _j_ 0) (33) 

"Vo r 



with 

'-V 77 ? (34} 

and 



cos6 = j (35) 

Curve 1 (D = 0.01) follows the equation 

y. = 1 - e -o.oo 5 *. cos x (36) 

which is very poorly damped. 
Curve 2 (D = 0.316) 

Oi -o.u8*. sm (0.99* + 0.141) (57) 

This curve is much better damped and is therefore plotted on a wider scale to 
follow individual vibrations. 
Curve 3 (D = 1.41) 

- = 1 - 1.41-e-- 707x sin (0.707* + 0.785) (38) 



254 W. J. ALBERSHEIM AND L. F. BROWN [j. S. M. P. E. 

The damping is so great that the ribbon overshoots only 6 per cent. 
(5.2) Aperiodic Damping. For D ^ 2 the transient decay is aperiodic accord- 
ing to 



yo n m \_m 

with 



and 



Curve 4 (D = 10) is drawn according to 

2- = 1 - l.Ole-O- 1 * + 
^o 

It is seen that due to overdamping the response is sluggish. From the transient 
as well as the steady-state point of view a damping factor between 1.2 and 1.4 is 
most desirable. 

(6) POWER CONSUMED BY FEEDBACK COIL 

The voltage delivered by the feedback amplifier is about proportional to fre- 
quency, and at resonance nearly enough to cancel the driving voltage. The cur- 
rent is inversely proportional to the undamped mechanical ribbon impedance; 
hence 

e F = -e N (43} 

i F = to(l - JV 2 ) (44} 

W F = -W driV e-N(l - N*} (45} 

This reaches a maximum at 



-JS 



(46} 



which for a valve strung to 10,000-cycle resonance corresponds to 5800 cycles. 
The maximum power consumed is 



W max = -WM*- -\'l ~ = 0.356 W*fc. (47} 

The feedback amplifier absorbs at most 8 db less than the low-frequency driving 
power of the light-valves. At this power level low distortion is not very essential 
because the second harmonic falls to the frequency 

2N m = 1.16 (48} 

which usually is beyond the transmission range of the recording and reproducing 
channels. 



Mar., 1942] STABILIZED FEEDBACK LlGHT- VALVE 255 

DISCUSSION 

MR. KELLOGG : In most of the applications of the feedback principle the pur- 
pose has been to provide linearity and constant input-output ratio over a wide 
frequency range. It is my understanding that the application which we have 
just heard described does not undertake this, but merely employs feedback to 
damp out a resonance peak. That being the case, I can see that certain other- 
wise difficult problems are avoided. Since only a narrow frequency range need be 
covered, transformers in the circuit are not a menace to stability, and the fact 
that a recording galvanometer is essentially a constant amplitude device, whereas 
the voltage fed back is proportional to velocity rather than amplitude, does not 
cause any embarrassment. Is this a correct interpretation? 

MR. ALBERSHEIM: Yes; the interpretation is correct. 

MR. REICHES: By the use of the Heavyside unit function, the authors have shown 
that with feedback it is possible to eliminate or greatly to minimize the occurrence of 
the transient described by the unit function. It is probably true that every piece of 
equipment built and used by us today is subject to the production of an impulse 
transient which can occur whenever the power-factor of the equipment is other 
than unit. This type of transient, in all probability, is a large contributing fac- 
tor to harshness and raspiness of speech in linear systems. Depending upon how 
far the feedback effect is carried, one would almost expect to hear the difference 
between two recordings, one with and one without the feedback light-valve ampli- 
fier. Sounds that are sharp in build-up, which may be found in words such as 
cook and book, should be improved. 

Possibly the term "impulse" transient should be better defined: It is a term 
sometimes used to describe the transient produced by a discontinuity in either the 
electrical or magnetic field, the transient being produced by the field which is 
double- valued at the same instant. The clashing of the light- valve ribbon may 
produce such a transient. 

However, it is further necessary to ask whether the other type of transient has 
been improved. It is sometimes called the "hyperbolic" transient, taking its name 
from the fact that the rate of growth of sound in general is a hyperbolic function 
of the form. This transient is largely determined by the phase-shift through the 
system between the various sine-waves forming the complex sound. 

Black points out that a feedback amplifier may have a phase-shift exactly equal 
to the feedback network +180; however, still considering the degenerative case, 
it seems possible that the phase-shift through the system may be aggravated. 
Offhand, the circuit shown by the authors may tend to approximate a linear phase- 
shift condition. I wonder whether any evidence exists showing that the condition 
due to non-linear phase-shift has been improved. 

MR. ALBERSHEIM: The two types of transients named by Mr. Reiches are in- 
terrelated just as are the "impulse" and the "steady-state" response of a system. 

To respond faithfully to an instantaneous impulse, a transmission circuit must 
have a steady-state frequency characteristic that is flat from zero to infinity. 
Generally, large amplitudes and slow decay of transient oscillations will corre- 
spond to sudden shifts of the steady-state amplitude and phase characteristics 
near the frequency of the transient disturbance. 

The steadying effect of feedback upon the light-valve has been demonstrated 
by recordings of steady-state square-wave signals as well as of single impulses. 



A CONSTANT-TORQUE FRICTION CLUTCH FOR FILM 

TAKE-UP* 



WILLIAM HOTINE** 



Summary. An analysis of f fictional take-up devices in general use discloses the 
reasons for their failure or improper operation. An improved friction clutch is 
described and analyzed, which embodies means for canceling the effect of a varying 
coefficient of friction encountered at variable slip speeds, thus maintaining a constant 
torque at the take-up spindle. Automatic declutching occurs when the load exceeds 
the amount of torque transmitted, thus protecting the film from excessive tension, and 
eliminating declutching levers and ratchets when rewinding. The novel construc- 
tion enables bidirectional operation, both in direction of rotation and in direction of 
transmission of torque. 



Important factors that should be considered in motion picture 
equipment design are reliability, efficiency, and protection of the 
film. The film take-up device, which is an important functional 
unit of all motion picture machinery, should naturally be designed 
and constructed with these factors in mind, as it is an integral part 
of the film-handling mechanism. Proper functioning of the take-up 
device enables the highest quality of performance of the associated 
machine, while poor take-up operation may not only adversely af- 
fect the performance, but may cause film damage. In considering a 
take-up design, the importance of reliable, safe operation should out- 
weigh any question of economy and simplicity. 

Let us examine the take-up problem. The film is leaving the ma- 
chine at a constant linear speed, and is wound on the take-up reel at 
a constantly increasing diameter, so that the rate of rotation of the 
take-up reel is constantly decreasing. During the take-up process, 
the film tension should not exceed a safe value. The film should be 
wound evenly and firmly on the reel, so the film tension should not 
fall below a safe minimum value. The take-up spindle, therefore, 
should be driven at a varying speed, and at a definite, limited torque. 

* Presented at the 1941 Fall Meeting at New York; received October 20, 
1941. 

** Rotovex Corp., New York, N. Y. 
256 



CLUTCH FOR FILM TAKE-UP 257 

The highest speed will occur when the film starts winding on the take- 
up reel hub, and the limit of torque will be that which produces a 
safe film tension at this point. In most motion picture machinery, 
the take-up spindle is driven at this speed and torque by means of a 
slipping frictional device, such as a belt slipping on a pulley, or a 
friction clutch. 

The slipping-belt take-up device is quite widely used and has the 
advantages of simplicity and economy. The torque transmitted to 
the take-up spindle is equal to the product of the pressure of the belt 
on the pulley surface, the radius of the pulley surface, and the co- 
efficient of friction between the belt and pulley. The pressure of the 
belt on the pulley depends on the belt tension, which is hard to main- 
tain at a uniform value. The area of contact between belt and pulley 
is small, resulting in high unit pressure which tends to wear or polish 
the pulley, changing the coefficient of friction. The only constant 
factor is the pulley radius. The slipping-belt take-up device is not 
a preferred method, from the standpoint of reliability, as the take-up 
torque may change at any time due to changes in the factors in the 
torque equation. 

The friction-clutch take-up device offers a more reasonable solu- 
tion to the problem, combining simplicity and economy with better 
stability of operation than the slipping belt. An ordinary friction 
clutch may be described as comprising a rotating driving member, a 
frictional element, and a rotating driven member which directly 
drives the take-up spindle. These elements are held in frictional en- 
gagement by the pressure of a spring. The torque transmitted to 
the take-up spindle is equal to the product of the mean radius of the 
members, the spring pressure, and the coefficient of friction of the 
members and the frictional element. Wear can be reduced by de- 
sign for small values of unit pressure on the frictional elements. The 
spring pressure can be easily regulated and will remain substantially 
constant, as will the coefficient of friction. The friction-clutch take- 
up device is to be preferred to the slipping belt, as reasonably reliable 
operation can be expected. The spring pressure, however, is the 
largest factor in the torque equation, and in a typical design the take- 
up torque will vary almost directly as the spring pressure, thus re- 
quiring a critical spring pressure adjustment. At the low spring 
pressures encountered, a slight lessening in spring pressure caused by 
wear of the clutch elements might easily reduce the take-up torque 
to an insufficient value. Careful readjustment of the spring pressure 



258 



W. HOTINE 



[J. S. M. P. E. 



is required to prevent an excessive take-up torque with resultant 
damaging film tension. It can therefore easily be seen that an im- 
provement in friction clutches in which there would be a reduction 
of the spring pressure factor in the torque equation, and in which the 
wear of the frictional element would have a smaller effect in reducing 
the spring pressure, would be desirable. 

The improved constant-torque clutch to be described embodies the 
above-mentioned desirable characteristics in its construction and de- 
sign. Fig. 1 is a cross-section in front elevation of a take-up device 
for use on a 16-mm projector, using this improved friction clutch. 
The supporting arm carries a sleeve shaft 24 rotating in a bearing, 
with driving pulley 22 fastened on one end of the sleeve, and driving 




FIG. 1. Cross-section of take-up device. 



clutch-disk 23 fastened on the other end. The driven shaft 27 has a 
thrust shoulder and take-up reel spindle on its left-hand end, and is 
free to revolve inside the sleeve. The round driving pin 28 extends 
radially from shaft 27, to engage the cam surface on the end of cam 
sleeve 26, which is free to revolve around shaft 27. The driven 
clutch-disk 35 is fastened to cam sleeve 26, and is held against fric- 
tion washer 32 by the spring. Adjustment of spring pressure is made 
by nut 31. 

In operation, a belt drives the pulley in a clockwise direction, 
viewed from the left-hand side, revolving the sleeve shaft and the 
driving clutch-disk, and imparting a torque to the driven clutch- 
disk via the friction washer. The torque of the driven clutch-disk 
is transmitted to the driven shaft by means of the driving-pin and 
the cam surface on the end of the cam. 



Mar., 1942] CLUTCH FOR FlLM TAKE-UP 259 

In the ensuing discussion the following nomenclature will be used : 

F = force of spring pressing on clutch-plate. 

/ = total force pressing clutch-plates together. 

f p = force between driving-pin and the wall of the cam. 

f a = component of f p in the axial direction. 

T = torque delivered by the clutch to the load. 

R = radius of gyration of clutch-plate. 

r = radius of effective point of contact between the driving-pin and face of 
cam. 

/3 = angle between a tangent to the cam surface and a diameter of the shaft. 

X = the coefficient of friction between the frictional element and a clutch- 
plate. 

It can be seen that the spring pressure is exerted against the driv- 
ing-pin in an axial direction, and part of the spring pressure is exerted 
in engaging the clutch-disks. The total force pressing the clutch- 
plates together is equal to the force of the spring minus the axial 
component of the force between the driving-pin and the cam surface. 
This relation may be expressed by 

/=*-/ (1) 

The axial force is the component along the axis of the total force 
f p between the pin and the surface of the cam, and therefore: 

fa = /, COS p (2) 

The torque delivered to the shaft is equal to the radius of the point 
at which the pin presses against the cam surface multiplied by the 
component of the force on the pin which is tangential to the shaft. 
This relation is expressed as 

T = rf p sin (3) 

If we divide 2 by 3 we obtain the relation between the axial force 
and the torque; thus 



If the value of f a given by 4 is substituted into 1, there results 

/ = F - 1^1 (5) 

The torque is equal also to the radius of gyration of the clutch- 
plate multiplied by the tangential force between the plates. The 



260 W. HOTINE [J. S. M. P. E. 

latter is equal to the force pressing the plates together multiplied by 
the coefficient of friction. Hence we may write 

T = R\f (6) 

If we eliminate/ from equations 5 and 6 there results 

T -- - (7) 



Equation 7 gives the torque delivered to the take-up spindle in 
terms of the various mechanical constants of the mechanism. 

The torque delivered by the clutch will be a constant for a given 
adjustment of the spring pressure, as shown by the above deri- 
vation. The torque can never exceed this constant value, as an 
additional load will automatically lessen the torque transmitted to 
the driven clutch-disk as the spring is compressed by the climbing 
motion of the driving-pin up the cam surface, increasing the axial 
force on the driving-pin. The film is thus protected at all times 
from a damaging tension. A constant torque at the take-up spindle 
means that the film tension will be inversely proportional to the 
radius of the film pile on the reel. If this radius increases from two 
to seven inches during the take-up operation, it follows that the 
film tension at the end will be only 30 per cent of that at the start. 
The ordinary clutch does not deliver a constant torque at a variable 
speed because the coefficient of friction decreases with increased 
slip-speed of the clutch faces. In winding film, when the film pile 
increases in diameter, thus slowing down the take-up reel and in- 
creasing the slip-speed of the clutch faces, the torque of the ordinary 
clutch decreases, resulting in a loosely wound film pile at the larger 
diameter. This condition is aggravated by the increased film weight 
at large diameters. If the spring pressure is adjusted for sufficient 
torque to overcome this condition, the film tension at the start of 
the winding is generally excessive, and is conducive to film damage. 
The improved clutch delivers a constant torque at variable speeds, 
because any change in the coefficient of friction substantially cancels 
itself out, and has little effect on the torque, as can be seen by in- 
spection of equation 7. This constant torque can not be exceeded, 
thus affording definite protection from excessive film tension under 
ah 1 conditions, and as the torque is maintained regardless of the film 
pile diameter, a tightly wound reel always results, even with film 



Mar., 1942] 



CLUTCH FOR FILM TAKE-UP 



261 



tensions ranging down to three-quarters of an ounce at the periphery 
of the reel. 

Fig. 2 graphically indicates the smaller variation in output torque 
of the improved clutch. Spring pressure in pounds is plotted against 
output torque in foot-pounds for the ordinary clutch, and for the 
improved clutch designed for the same output torque. The smaller 
variations in torque with variations in spring pressure shown in the 
curve for the improved clutch, together with the much larger spring 
pressures required for a given value of torque, indicate a more stable 
condition which would tend to maintain an adjustment and produce 
more reliable take-up operation. The angle /? of the cam surface 
determines the slope of the curve for the improved clutch. As the 



TORQUE 



SPRING PRESSURE 
ORDINARY CLUTCH ROTOVEX CLUTCH 

FIG. 2. Variation in output torque. 

cam surface may be constructed to give bidirectional operation, 
different angles on the two sides of the cam will enable different 
torques to be transmitted with the same spring pressure, depending 
upon the direction of rotation. 

Other features of the improved clutch are: Automatic declutch- 
ing of the plates when the take-up spindle is driven in reverse, as 
when rewinding; ratchets and declutching levers can be dispensed 
with, further simplifying construction and operation of the film take- 
up mechanism. The direction of transmission of torque can be 
reversed with identical operation; that is, the power can be applied 
to what was formerly the driven member, in adapting the construc- 
tion to suit other purposes such as brake mechanisms. 

In conclusion, the writer wishes to acknowledge the able assistance 
of Dr. Charles B. Aiken in the preparation of this paper. 



RECENT DEVELOPMENTS IN PROJECTION MECHANISM 

DESIGN* 



EWALD BOECKING AND L. W. DAVEE** 



Summary. A 35-mm motion picture projector mechanism is described. A 
theoretical discussion of the operation of single and double-shutter mechanisms shows 
the advantages of each and covers points that should be considered for improving screen 
reproduction from mechanisms generally. 

With the passing of each year, the requirements of the motion 
picture industry and the projection of the picture in particular have 
become more and more exacting. These improvements include such 
items as greater accuracy in projection, increased operating effi- 
ciency, lower maintenance, and longer life. Thus the demands upon 
the projection mechanisms have become increasingly severe. 

The acceptance by the industry of certain standards which were 
dictated by previous use, rather than a redesign based upon an 
engineering foundation, has resulted in the fixing of physical dimen- 
sions and shaft speeds which for commercial reasons can not be 
changed without the cooperation of the industry. If certain of these 
standards could be changed marked improvement in mechanism 
design could be effected. 

These design features include the height of the optical centerline 
of the projection mechanism above the separation line between the 
sound-head and the projection mechanism ; the film distance between 
the optical centerline of the picture and the optical centerline of the 
sound-track; the projection of 24 frames per second; the speed of 
the shutter and intermittent cam shafts of 1440 rpm ; and the speed 
of the main drive-shaft which couples the mechanism to the sound- 
head. 

Having these standards already in existence it becomes an engi- 
neering necessity to design any new projection equipment to fulfill 
those requirements plus those mechanical and operating features 

* Presented at the 1941 Fall Meeting at New York, N. Y.; received October 1, 
1941. 

** Century Projector Corp., New York, N. Y. 
262 



PROJECTION DESIGN DEVELOPMENTS 



263 



desired in the final product. As in the design of any scientific or 
mechanical device the stability and inherent durability of the product 
must first begin with perfection in the basic design and it must be 
built on a foundation of engineering knowledge proved by practical 
operating experience. 

We shall begin our discussion with the main supporting frame. 
A mechanism involving rotating shafts, meshing gears, and accurate 
optical alignments must be assembled on a firm foundation free from 




FIG. 1. Rear view of single-shutter mechanism. 

the dangers of warping and guarded against shaft alignments when 
subjected to relatively severe shocks which necessarily occur, at 
times, in shipment or during installation. In Fig. 1 we see a center 
frame entirely enclosed in a cast case made in one piece. The center 
frame is therefore substantially supported on all sides, making any 
displacement of critical alignments virtually impossible. With the 
exception of the upper and lower sprocket-shafts, all the other 
shafts are supported by brackets attached to the main frame. In 
this design each shaft is placed in perfect alignment without the 
necessity of using alignment reamers, gauges, or other special tools. 



264 E. BOECKING AND L. W. DAVEE [J. S. M. P. E. 

Gear Design. The fundamentals of gear design are well known. 
Having definite requirements regarding load, ratios, shaft speeds, 
and operating functions, there is almost no choice in the selection of 
the gear design to accomplish efficiently the work to be done. Many 
gear designs are selected, however, without first studying the factors 
governing their operation and taking advantage of the type of gearing 
which best suits the purpose. 

Any pair of gears which are properly designed for pitch diameter, 
shape of gear teeth, pressure angles, etc., must, for efficient, quiet 
operation run on shaft centers dictated from the basic design rather 
than from the trial-and-error methods sometimes followed by a num- 
ber of assembly men. The prevalent use of grinding compound in 
fitting gears, to quiet their operation is in general due to the inac- 
curacies of manufacture or the failure of the mountings used to posi- 
tion the shaft centers correctly. These errors cause tooth interference 
which can at best be eliminated only partially, and always result in 
inefficient gear-trains. If grinding compounds are used it is practically 
impossible to remove the cutting abrasives entirely and they con- 
tinue to cause excessive wear throughout the life of the mechanism. 

There are no grinding compounds used in the assembly of the gear- 
train shown because each pair of gears has been designed correctly 
for the purpose for which it is used, and the shaft centers are fixed 
exactly by the mounting bracket design coupled with the use of ball- 
bearings insuring perfect shaft centering. 

Bearings. The bearings in this mechanism are of two types. They 
have been engineered to give long, trouble-free service in the par- 
ticular positions in which they are used. 

Except in the slow-speed upper and lower sprocket-shafts, sealed- 
for-life ball-bearings have been adopted for all the shafts in the 
mechanism. There is much misinformation and confusion regarding 
the use of ball-bearings in the reduction of wear and the maintenance 
of accuracy in machine design, and it is therefore desirable to review 
some of the facts regarding the operation of these bearings. 

Friction is a phenomenon that works for both good and evil. 
The effect of friction in a projection mechanism can cause only one 
result, increased driving power with a consequent increase in reflec- 
tion of gear disturbances transmitted to the sound-head. Mechani- 
cally, friction causes the generation of heat and the increase of wear. 
Sliding friction is always present in plain bearings and increases or 
decreases depending upon the type and kind of lubricant used, not 



Mar., 1942] PROJECTION DESIGN DEVELOPMENTS 265 

the amount. Higher-speed shafts naturally generate more heat and 
increase the amount of wear. 

In order to eliminate friction ball-bearings were developed, and 
their advantages appeared so promising that in Europe, and then in 
America, the most exhaustive scientific research was instituted to 
determine the fundamental principles underlying their theory. The 
ball-bearing is today, therefore, distinguished from all other forms 
of anti-friction in having a thoroughly sound background of scientific 
principles governing its design. The load-carrying capacity and ex- 
pected life are not merely matters of experiment but are subject to as 
accurate mathematical determination as is the strength of a steel 
bridge. 

Friction in any bearing is very largely shown by the amount of 
wear that takes place. In a ball-bearing properly sealed against 
dirt, wear for all practical purposes can not be measured even after 
long, hard service. Ball-bearings are the only type of anti-friction 
bearing that can be successfully sealed and lubricated for life. Evi- 
dence of these long-life characteristics is found in the thousands of 
sealed-for-life ball-bearings that have run for many years without the 
least attention. 

Ball-bearings do not suddenly break down for no apparent reason, 
nor do they wear out in the sense of loss of dimensions or accuracy 
of positioning characteristics. In a ball-bearing running under load 
the balls and raceways are subject to continuous repetition of stresses 
which, after long and carefree service, begin to show the effect of 
structural fatigue. This in no way impairs the usefulness or operat- 
ing qualities of the structure until after an actual breaking down of the 
material occurs. This is not a sudden occurrence and can be predicted 
with remarkable certainty. All other causes of failure are premature 
and can be prevented. 

A good many years ago the principle of preloading bearings to 
reduce deflection within the bearing, and therefore increase the 
rigidity of the shaft support, was developed. The effect of preload- 
ing, however, which bears a most direct relation to wear, and which 
is frequently passed unnoticed by machine designers, is the fact that 
correctly applied preloading increases the useful life of ball-bearings 
and in so doing eliminates the noise usually associated with their 
use. 

Resistance to deflection as a result of preloading has a most impor- 
tant influence upon machine life in that it assures definite maintenance 



266 E. BOECKING AND L. W. DAVEE [j. S. M. P. E. 

of accuracy limits within which machine parts are designed to operate 
for best performance and endurance. 

Thus if mating gears are held very closely to the pitch-line relation- 
ship for which they are generated minimum gear wear will occur and 
maximum life and freedom from noise will result. Special impor- 
tance attaches to the ability of ball-bearings to achieve this result, 
for it is well known that if carefully located gears are permitted to 
change position gradually through frictional wear in the supporting 
bearings, and the gears have worn out of pitch, no amount of sub- 
sequent bearing alignment can restore them to their original ef- 
ficiency and quietness. 

Where closely and accurately fitted bearings are required, as in a 
projector mechanism, considerable preliminary adjusting has always 
been necessary before gears and shafts would operate properly. In 
addition, these bearings when worn in service, usually can not be 
replaced without delay. 

With the standardization and closely held limits permitted in ball- 
bearing manufacture it is now possible to eliminate all handwork and 
replace bearings without special skill, with the assurance of main- 
taining the original accuracy of the mechanism. 

Primitive is the idea that ball-bearings should be provided with 
a means of relubrication. This simply means that either too much 
grease, unclean grease, or the wrong kind will sooner or later be added. 
It likewise means that either the seals or the original lubricant is not 
good enough. The bearings adopted in this mechanism are therefore 
sealed for life, and never require maintenance or relubrication of any 
kind. They are bearings of proved performance, lubricated and sealed 
to protect and preserve their accuracy by the best means that the wid- 
est experience in the industry can devise. There are at present over 
10,000,000 of these self-sealed bearings now in use giving trouble- 
free, dependable service. 

Each bearing is mounted in a bracket attached to the main frame 
by means of carefully machined surfaces and heavy mounting screws. 
In this method of design entire shaft and bearing assemblies may be 
interchanged without the necessity of hand-fitting or shaft align- 
ment. The design also precludes any possibility of bind-up. 

The upper and lower sprocket-shafts have been designed using oil- 
less type bearings. In shafts such as these, ball-bearings are not war- 
ranted for several reasons. The speed of rotation is relatively slow, 
and therefore they are subject to very little wear. This type of 



Mar., 1942] PROJECTION DESIGN DEVELOPMENTS 267 

bearing should properly be called "oil-plus" instead of oil-less. They 
are made of 99.8 per cent pure powdered bronze, compacted or 
squeezed together under hydraulic pressure, forming a homogeneous 
capillary structure. These fine particles are then permanently 
joined together by a heat-treating process. After the heat treatment, 
the metal is full of tiny capillary pores too small to see without a 
powerful magnifying glass but able to soak up oil like a sponge. It 
is then impregnated to 1 / 3 of its volume with pure acid-free oil. 

The moment the shaft begins to rotate in this bearing the pressure 
and movement instantly cause a film of oil to be formed between the 
moving parts. The bearings therefore do not require any additional 
lubrication over long periods of service. Tests recently completed 
indicate that after 1,419,824,000 revolutions of a shaft in these bear- 
ings, without a stop and without one added drop of lubrication, 
neither the shaft nor the bearing showed signs of wear. At 360 rpm, 
the speed of the upper and lower take-up shafts, one can make his own 
calculation as to their probable life. Channeling of shafts and bear- 
ings has thus been eliminated with an assurance of long service free 
of bind-up. It therefore can be seen that practically no lubrication 
is required that will come into contact with the film during projection, 
causing oil marks and the collection of dirt. 

The main drive-shaft, running in its sealed-for-life ball-bearings, 
is designed so that its centerline distance from the picture aperture 
is the same as has been adopted through custom. Any regular sound- 
head coupling gear and shaft will fit properly on this driving-shaft, 
but the small 17- tooth pinion usually used as a mechanism drive is 
not used as such. This small gear is used only as a clutch member 
with or without teeth. The other part of the clutch is the main 
drive-sleeve of the mechanism. The whole clutch assembly clamps 
together as a complete unit so that the power supplied from the sound- 
head drives directly to the vertical shaft through the main drive-gear. 

Gears. Each gear in the mechanism has been designed to do a 
particular job. There are no gears in the train that merely transfer 
motion from one part of the mechanism to some other part. As a 
result fewer gears are used, resulting in a consequent reduction in 
friction and wear. 

The sound-head drives directly to the vertical shaft through a gear 
'specially designed for the smooth efficient transmission of power. 
From this vertical shaft are then driven all the other operating parts 
of the mechanism. The advantages of this system are apparent. 



268 



E. BOECKING AND L. W. DAVEE 



[J. S. M. P. E. 



Backlash, the cause of jumping gear-trains, and the introduction of 
excessive travel-ghost have been reduced to a minimum. Only two 
gear contacts exist between the intermittent cam and the shutter 
blades. By the same token only two gear -contacts exist between the 
main drive and the shutter or the intermittent. Such design has 
contributed much toward making this the quiet, stable, and efficient 
mechanism which it is. 

Scientifically designed, hunting-tooth spiral gears throughout 
assure quiet operation and an absence of reflected, varying loads back 
to the sound-head gears. All the steel gears are heat-treated and 
hardened to contribute to longer life with a consequent reduction in 




; (D 

FIG. 2. "Century" projector intermittent movement. 

the possibility of developed backlash. All gears have a face of l / 2 
inch or more and are meshed steel against fiber. The fiber gears are 
oil-impregnated so that only that amount of lubrication need be 
supplied to compensate for losses due to evaporation, etc. 

Intermittent Movement. The intermittent movement is an out- 
standing example of precision workmanship. It has been designed 
with four bearings, two on the cam-shaft and two on the sprocket- 
shaft. Perfect alignment is therefore maintained between the two. 

This is a direct-drive movement having no gears inside the case. 
The flywheel is mounted directly on the cam-shaft with no gears be- 
tween it and the cam. The flywheel, of course, supplies the power 
required to move the starwheel, sprocket, and film. Without the 
flywheel this intermittent load would reflect back into the gear-train 



Mar., 1942] PROJECTION DESIGN DEVELOPMENTS 269 

causing considerable disturbance. There being no gears between 
this intermittent load and the flywheel driving-power, the noise of 
operation is reduced and the operation of the starwheel made more 
positive. 

The four bearings are supported in heavy arms and are automati- 
cally supplied with lubrication through scientifically designed oil- 
channels. The outer bearings are of the same oil-less design as is 
used on the upper and lower sprocket-shafts and, in addition, are 
supplied with oil which, as needed, filters through the capillary tubes 
of the bearings. These tubes form an extremely fine filter for the 
oil and allow only that amount required for satisfactory lubrication 
to reach the shaft. 

The intermittent sprocket is designed in accordance with the ac- 
cepted standards of such sprockets and is heat-treated, hardened, 
and ground to YIO.OOO of an inch, as are also the cam, cam-pin, star- 
wheel, and shafts. 

A micrometer adjustment is provided so that the distance between 
the cam and the starwheel may be changed while the mechanism is 
running. With this adjustment the intermitted may be set for the 
best screen results together with quietness of operation. 

The intermittent pad-shoe assembly is mounted directly on the 
case of the intermittent, and is designed so that each shoe is self- 
aligning to the sprocket with pressure applied to each shoe equally. 
With this type of construction, uniform film motion through the 
trap-and-gate assembly is assured with a freedom from unequal 
wear and film buckle heretofore unattained. 

The intermittent movement is easily removed from the operating 
side of the mechanism so that accidental damage to the sprocket is 
minimized and the removal of gears rendered unnecessary. Only 
the flywheel has to be removed. The movement may be replaced 
in the mechanism in perfect time with the shutters without the 
necessity of marking gears or dowel-pin arrangements. Timing the 
shutters with the intermittent is also accomplished without the use of 
aligning tools and without removing any of the operating parts such 
as lenses, etc. 

The Operating Side. A full- width symmetrical door is provided 
to allow ready access to the operating side of the mechanism and 
gives plenty of room for threading or servicing. All spring tension 
in pad rollers and gate is obtained with coiled springs under com- 
pression, so that there is minimum danger of spring breakage. This 



270 



E. BOECKING AND L. W. DAVEE 



[J. S. M. P. E. 



construction also reduces the possibility of accidentally changing 
tension adjustments by bending the springs, catching of cleaning- 
cloths, etc. 

The pad roller construction is a complete operating unit. The 
tension is obtained from a construction employing two coiled springs 
under compression, locating two steel balls into recesses of the hard- 
ened steel hub of the pad-roller shaft. With this type of self-con- 




FIG. 3. Operating side of mechanism. 

tained arrangement the complete pad-roller assembly may be removed 
from the mechanism for replacement, cleaning, or adjustment. 

In addition to being complete in its working parts, the strippers 
are included as an integral part of the pad-roller hub. Thus the 
stripper is always in adjustment and is automatically positioned 
when the pad-rollers are closed. This feature makes it most con- 
venient for sprocket replacement. With the pad-rollers in open 
position, the sprockets may be replaced without removing or re- 
adjusting any other parts of the mechanism. 



Mar., 1942] PROJECTION DESIGN DEVELOPMENTS 271 

In threading the film, the pad-rollers are opened and provide a 
gauge for setting the loop sizes above the film-trap and below the 
intermittent sprocket. The film is threaded tight over both pad- 
rollers, and the loop sizes are automatically set when the rollers are 
closed. 

The film-trap and gate have been designed to fill not only the tech- 
nical requirements for perfect picture reproduction, but also to pro- 
vide the most desirable features from an operating standpoint. The 
film-trap is a complete assembly readily removable from the mecha- 
nism for inspection and exactly interchangeable without readjust- 
ment. It includes interchangeable film-shoes and studio guides, heat- 
treated, hardened, and ground. The upper guide-roller is designed 
so that the guiding face is integral with the shaft on which it is 
rotating, and is ground on the same centers on which it is mounted in 
the film-trap. This construction assures perfect accuracy of the 
guiding face and eliminates the side weave that will occur if there is 
any eccentricity of the surface. 

The film-trap is mounted to the main frame with three large 
screws, and is held in accurate alignment to the optical centerline 
by means of the large ground pad, which forms the base of the film- 
trap, and two dowel pins so located that the complete trap assembly 
may be moved forward and back parallel to the optical centerline. 
This adjustment is provided so that perfect alignment may be ob- 
tained between the trap shoes and the faces of the intermittent 
sprocket. 

A number of experiments have been made to determine the best 
relation of the alignment between the trap shoes and the sprocket. 
From jump measurements which were made, it seems desirable to 
set the shoes slightly ahead of the sprocket face so that effectively 
the film position under tension is exactly parallel to the face of the 
trap shoes. 

The tops of the trap shoes are curved back away from the flat 
surfaces of the shoes, allowing the upper loop to be made in back of 
the centerline of the film in the film-trap. The reason for this will 
be discussed under the explanation of the film-gate operation. 

The lens-holder and film-gate mounting is a complete unit, readily 
removable from the mechanism. As in the case of the film-trap it 
also is mounted to the main frame by means of a large accurately 
ground pad and held securely in position with large screws. The 
complete assembly is also located with two dowel pins so designed 



272 E. BOECKING AND L. W. DAVEE [J. S. M. P. E. 

that it may be moved toward the film-trap or away from it. With the 
film-trap properly located with respect to the intermittent sprocket 
the gate may then be set to its proper distance from the trap, giving 
an optimal operating condition for each unit. 

The lens-mount is designed for "half -size" lenses, and may be used 
with any lens by means of adapters. 

The opening and closing of the gate as well as its accurate align- 
ment with the trap is accomplished by means of a large sliding tube 
mounted inside the frame of the lens mount. Both the inner sur- 
face of the frame and the outer surface of the tube have been ac- 
curately ground to provide a smooth, substantial mounting for the 
gate. When the gate opens and closes, it always moves on the 
optical centerline and can by no bending or displacement of the 
mounting fail to register in perfect position with the trap. 

The gate opening and closing knob located in the center of the 
mechanism contains all the tension springs, locks, etc., associated 
with the gate movement. Here again there are no flat springs to 
get out of adjustment. 

A cam action inside the knob securely locks the gate in the closed 
position so that it is impossible to open it accidentally even though 
the spring tension were to fail. Two open positions of the gate 
provide for normal gate opening for threading, and the second and 
larger opening, which is about one full inch, gives plenty of room for 
trap-shoes and gate-pad inspection and cleaning, if necessary, without 
removing them from the mechanism. The gate is a real advance in 
design and was adopted after many models and designs had been 
discarded. 

The gate-pads are mounted on the gate-plate, which is flat and ac- 
curately ground on both front and rear surfaces, by means of equaliz- 
ing levers which insure exactly the same film tension on both sides of 
the gate. The tension is obtained by means of two large coiled 
springs under compression, and is adjustable from zero tension to a 
safe maximum. A safety stop is provided so that there is no danger 
of locking the film during projection. 

At the top of the gate are mounted two guiding shoes, called upper 
loop stabilizers. These stabilizers rest against the film above the 
gate-pads, giving the film a slight bend backward from the trap- 
shoes. The result of this feature has been to prevent the reflection of 
disturbances to the film at the aperture caused by the changing upper 
loop. 



Mar., 1942] 



PROJECTION DESIGN DEVELOPMENTS 



273 



Experiment has shown that no matter how long the film-trap was 
made or how securely the edges of the film were held, deflections of 
the film at the aperture could not be prevented unless a method such 
as described above was used. These deflections occurred as the upper 
loop changed its size during the projection of the picture while the 
film was stationary in the aperture. The resulting screen appearance 
was the same as diffusion in the lenses. The sharper focus of the pic- 
ture on the screen is quite obvious with this new gate design. 

The Shutters. Theoretically if one edge of a 90-degree shutter 
intercepts a beam of light on the axis of the optical system at the 



ze OF Donate suurrct BLAOC 

ACTUAL tilt OF OOUBLC SHUTTCK ILA 




AT 

BEGINNING OF PULL 
DOWN 



FIG. 4. Diagram of "Century" single and double-shutter interception. 



time of engagement of the intermittent-cam pin with the starwheel, 
the trailing edge will leave this beam at the time the intermittent- 
cam pin leaves the starwheel. This relation is true with a Geneva- 
type 90-degree movement. It is the practice for almost all projector 
manufacturers to use this type of movement. (Fig. 4.) 

However, the beam of light intercepted by the shutter is in reality 
of finite size and becomes appreciable in the operation of the pro- 
jector mechanism. It is therefore no longer possible to consider 
shutter operation on the basis of its theoretical optimal operation 
and we must investigate the factors governing its effect on the picture 
reproduction. 



274 E. BOECKING AND L. W. DAVEE [j. S. M. P. E. 

One blade of the shutter, called the flicker blade, does nothing 
more than to increase the number of pulsations of light that the eye 
views until the persistence of vision gives the impression that a 
steady light is being projected. The higher the screen illumination 
the more critical the eye becomes of light variations. Therefore in 
the following discussion of shutter design it will become obvious that 
the higher the screen illumination the greater becomes our inter- 
ference factors. 

The average single rear-shutter mechanism requires a 95 to a 110- 
degree cut-off blade for reasonably good screen performance. This 
operation is dependent, of course, to some extent upon the speed of 
the projection lens being used. In our discussion we shall assume 
one lens size as a constant and relate all the discussion to it. These 
mechanisms referred to are the type which usually employ so-called 
90-degree shutters. The average double-shutter mechanisms using 
double rear shutters according to the usual practice require cut-off 
blades of 70 to 75 degrees for optimal performance. 

An investigation of the shutter problem has resulted in the design 
of a shutter combination that will give optimal operating results 
consistent with the physical dimensions that were decided upon. A 
higher-efficiency mechanism could be designed but its size would 
make it inconvenient for the average projection room. 

Seven fundamental factors were considered in the design of the 
shutter on this mechanism: 

(1} Heat on the aperture. 

(2} Total screen illumination or shutter efficiency. 

(5) Amount of travel-ghost allowed. 

(4) Size of the light-spot at the point of cutting. 

(5) Position of the shutter in the light-beam. 

(6) The peripheral speed of the shutter-blade. 

(7) Mechanical strength and wearing qualities. 

A balance of all these factors must be made and the engineering 
considerations carefully weighed. 

The total heat at the aperture, regardless of the cooling methods 
employed, obviously dictates the use of rear shutters, either single or 
double. The heat at the aperture should also be in proportion to the 
amount of light being transmitted. 

The size of the beam of light intercepted governs the time that a 
shutter blade at any peripheral speed requires to cut off the light 



Mar., 1942] 



PROJECTION DESIGN DEVELOPMENTS 



275 



completely. The peripheral speed of the shutter-blade also governs 
the time required to cut a beam of light of any given size. 

If we consider a light-beam from the lamp-house mirror having an 
angle of about 20 degrees, the size of the spot 4 inches from the aper- 
ture will be about 2Vs inches in diameter. At a distance from the 
aperture of 5 inches this beam will be about 2 1 / 2 inches, which 
amounts to an increase of I7 l /z per cent travel time of the shutter- 
blade. If the distance is increased to 6 inches the beam will increase 
to about 2 15 /ie inches or an increase in travel time of 38 per cent. 



SHUTTER BLADES 




FIG. 5. Diagram of light-beam interception. 

The peripheral speed of the shutter, which has a rotary speed of 
1440 rpm, will be governed by the distance from the center of the 
shutter-shaft to the center of the beam of light. The arc of the 
shutter-blade at the center of the spot is the longest distance the 
shutters have to travel. 

Assume that the light-beam where the shutter is located is 2 1 / 8 
inches in diameter and the distance from the center of the shutter- 
shaft to the center of the light-beam is 4 3 / 8 inches. The shutter- 
blade then has to travel about 28 degrees to cover the spot, which 
amounts to about 7 3 /4 per cent of one shutter revolution. 

If the distance from the center of the shutter-shaft to the center 
of the light-beam is 3 7 / 8 inches and the spot of light is increased to 
2*/2 inches, the cutting angle will become 38 degrees or about 10 1 /2 



276 



E. BOECKING AND L. W. DAVEE 



[J. S. M. P. E. 



per cent of one shutter revolution. Should the spot of light increase 
to 2 16 /ie inches and the distance from the center of the shutter-shaft 
decrease to 3 1 /z inches, then the cutting angle will increase to 51 
degrees or about 14 per cent of one shutter revolution. 

These angles of movement are called dead angles, and are a loss of 
picture projection time which can not be recovered after the mecha- 
nism has been designed and built, because of the physical limita- 
tions imposed by the design. 




FIG. 6. Diagram of intercepting angle. 



Considering a balanced shutter, a design having the cut-off speci- 
fications illustrated in case C (Fig. 5) would intercept the light-beam 
in 26 to 45 per cent less time than a design having the features 
illustrated in cases A or B. A correct conception of this relation 
would be as follows: if the conditions of picture reproduction in 
cases A and B are satisfactory for a given light-intensity on the screen, 
then the operation of the mechanism in case C would be satisfactory 
with shutter-blades of 10 to 23 degrees smaller angle. 



Mar., 1942] PROJECTION DESIGN DEVELOPMENTS 277 

Now let us consider the next step in the shutter design. (Fig. 4.) 
At a projection rate of 24 frames per second and using a Geneva-type 
intermittent sprocket pull-down, the starwheel movement time is 
theoretically 1 / 96 of a second, which corresponds to 90 degrees or the 
fourth part of a complete shutter revolution. 

Theoretically a film frame should not be moved before the light 
has been completely intercepted, and it should stop moving before 
the light is again transmitted. This means that a shutter should 
have a 90-degree cut-off blade plus the angle required to intercept 
the beam, which in our case is 28 degrees, or a total of 118 degrees. 

In case B having a 38-degree cut-off angle the total blade would 
be 128 degrees. In case A having a 51-degree cut-off angle the total 
blade would be 141 degrees. 

In practice, however, the cut-off angle of the blade can be made 
smaller than the angles shown and yet give good picture reproduction, 
so there must be some explanation for it. 

As the cam-pin enters the starwheel slot it will travel about 9 
degrees before the film starts to move. About 9 degrees before the 
cam-pin leaves the starwheel slot the film stops moving. During 
this 9-degree cam-pin travel the film does not move; therefore the 
shutter-blade can be designed with 18 degrees less angle than cal- 
culated, making it about 100 degrees. 

The difference between this 100-degree theoretical shutter-blade 
and the angle of shutter-blade actually used can not be explained 
mathematically or mechanically. There are, therefore, other factors 
governing the shutter operation that must be considered, and which 
will influence the quality of reproduction from any mechanism de- 
pending upon how far the particular designer wishes to go. 

From a number of experiments it has been determined that a 
motion picture would not look real if it were not for the inefficiency 
of the human eye, commonly known as persistence of vision, which 
allows the eye to dissolve one still picture into another one in a 
fraction of a second. This so-called identification deceiving act 
may need from l / 3 to l / 6 of a second. It seems that it has no effect 
upon reducing the size of the shutter-blade, however, and depends 
upon the frequency of the change and the amount of displacement 
occurring from one picture to another. 

There is another factor about which we are most concerned, and 
this is the time the eye requires to recognize a picture after a dark 
period. This time factor is influenced very greatly by the brightness 



278 E. B DECKING AND L. W. DAVEE [J. S. M. P. E. 

of the picture during the transmission period, and appears to be 
about VTSO to Vieoo of a second with the usual screen illuminations 
used in practice. This would account for the 12-degree smaller 
angle that can be used on the shutter, bringing the total single-shutter 
cut-off blade down to 88 degrees. As the angle is decreased a conse- 
quent increase in screen illumination is obtained which decreases the 
time of perception of the eye to travel-ghost. Therefore a balance 
must be obtained at which the degradation of picture quality, because 
of these effects, remains within satisfactory limits. 

In projector mechanism design using greater distances from the 
aperture to the shutter-blades and smaller distances from the center 
of the shutter-shaft to the center of the light-beam, these effects 
become greater and greater. With the mechanism design being dis- 
cussed the cut-off blade angles were selected on the basis of picture 
quality and freedom from travel-ghost with its attendant picture 
degradation, giving higher screen illumination plus sharper defini- 
tion than has heretofore been possible. 

Another design feature that has to be considered is the direction 
of the shutter interception with respect to the light-beam in other 
words, the position of the center of the shutter-shaft with respect to 
the optical center of the mechanism. (Fig. 6.) 

If the area of the light-beam were circular it would make no dif- 
ference where the shutter was placed; but as the area changes in 
shape the farther the interception point is from the aperture, this 
angle becomes of considerable importance. (Fig. 6.) 

Locating the shutter-shaft center in a horizontal direction about 
4 3 /s inches from the light-beam center, the angle of interception as 
explained before is about 28 degrees. If the shutter-shaft is located 
above or below the light-beam center, by using axial intercepting 
blades they would have to be 31 degrees wide or an increase of about 
11 per cent in cut-off time. 

Placing the shutter-shaft center about I 1 / 2 inches above or below 
the light-beam center and 3 3 /4 inches away, the intercepting area 
would be increased to 35 degrees or 25 per cent. Intercepting the 
same spot with a tangential blade, the intercepting angle increases 
still further to 37 degrees or 32 per cent over the original 28 degrees. 

Intercepting a light-beam on an angle of 45 degrees and keeping 
the same shaft center and beam distance as in case C (Fig. 6) which 
required 28 degrees, an increase in interception time of about 11 per 
cent would occur; but if, in addition, tangential blades are used to 
intercept the beam, with the edges arranged to be horizontal at the 



Mar., 1942] 



PROJECTION DESIGN DEVELOPMENTS 



279 



middle of the aperture, the angle again increases to 48 degrees or 71 
per cent above the design of case C. 

The design of the double-shutter mechanism is such that both 
shutters are located at the rear of the mechanism in order that the 
heat may be reduced at the aperture in accordance with the total 
light being transmitted. Each shutter is driven independently so 
that the driving load is divided between two sets of gears. Each 
shutter is mounted on its own ball-bearings; therefore there is no 




FIG. 7. Rear view of double-shutter mechanism. 



sliding friction between them and each shaft speed is the same 1440 
rpm as in the single-shutter unit. 

In the double-shutter mechanism each blade has to travel only 
halfway through the light-beam for complete coverage of the light. 
This amounts to a reduction of this cut-off period of 50 per cent. 
Thus the double-shutter mechanism, in addition to the gain in light- 
beam interception time "of 26 to 45 per cent as explained for the 
single-shutter mechanism, cuts this time in half so that a gain of 63 
to 73 per cent is realized. Following the same line of reasoning as 
in the case of the 88-degree single-shutter mechanism the same facts 
apply to the double shutter. (Fig. 4.) 



280 E. B DECKING AND L. W. DAVEE 

Therefore, if a 118-degree single shutter can be operated with 30 
degrees less blade than is theoretically possible, then a theoretical 
90-degree double-shutter blade can be operated at 60 degrees pro- 
vided that the same amount of light is being projected on the screen. 
Due to the increase of about 19 per cent in the projected light and the 
reduction of 50 per cent in the cut-off period, 7 degrees or 25 per cent 
of the intercepting spot angle has to be added on each blade to bring 
the amount of travel-ghost content to the same relative value as on 
an 88-degree single shutter. Therefore a theoretical 90-degree double 
shutter can be operated at 67 degrees in a light-beam position C, at 
70 degrees in a position B, and at 73 degrees in position A. 

The light efficiency of the Century single-shutter and double- 
shutter mechanisms represents real values and with proper methods 
of measurements may be checked by illumination measurements 
made in any theater. The single-shutter mechanism has 88-degree 
blades with 92-degree openings, with an efficiency, of 51 per cent. 

The double-shutter mechanism has 67-degree intercepting blades 
running in opposite directions. The efficiency calculation differs 
slightly from that of a single shutter and equals 59 per cent in a light- 
beam position C, 56 per cent in a position B, and 53 per cent in a posi- 
tion A . The double-shutter mechanism CC will deliver approximately 
16 per cent more light than the single-shutter mechanism C. This 
amount, compared to a single-shutter mechanism using 95 to 110- 
degree shutter-blades, results in an increase in light from 25 to 51 
per cent. 

Of course, in the determination of shutter-blade sizes there is one 
very important consideration that should be mentioned. In the 
design of the gear-train between the shutters and in the intermittent 
movement, each pair of gears will contribute a certain amount of 
backlash which must be compensated for in additional degrees of 
covering. We believe that the design used in the present Century 
mechanism has as few gears as is possible, considering the speed of 
rotation and the change of motional direction between the inter- 
mittent and the shutters. Whatever advantage there may be in 
these mechanisms from this factor has been used in improving the 
appearance of the picture, rather than in reducing the shutter angles, 
and to give the additional illumination which could be obtained. 
We feel, therefore, that the design of these mechanisms has con- 
tributed much to the design standards of this industry. 



REPORT OF THE STUDIO LIGHTING COMMITTEE 



Summary. Increasing interest in exposure meters is noted, with a concerted effort 
on the part of one manufacturer to provide studios with improved specially designed 
meters for each studio's individual need. "Arc" and "Inkie" equipments are being 
used more interchangeably on black-and-white and color sets as their respective design 
fits the cinematographers needs. 



In its endeavor to keep the Society informed of new and immedi- 
ately anticipated development in technic and equipment for photo- 
graphic lighting, the Committee submits this brief report. The re- 
port is necessarily brief since there have been no developments in 
new lighting equipment, and the technic has not changed appreciably 
since the last report was made. 

The use of exposure meters seems to be growing, and a newly de- 
signed meter is now being introduced into active use in the studios. 
The designer of this meter has approached his problem with the idea 
that scene illumination is a dual problem for the cinematographer, 
in which the elements of illumination balance and prevailing illumi- 
nation level must both be considered. The illumination balance is an 
artistic problem to be handled by the individual cinematographer's 
unique talent and experience in this field. The evaluation of illumi- 
nation level is a technical problem. The new meter is designed to 
accomplish the task with a maximum of accuracy, and a great saving 
of time in the lighting set-up. It utilizes a translucent hemisphere 
as a means of collecting light; is calibrated in lens// values instead of 
in foot-candles; and the meters are individually calibrated to meet 
the requirements of different studio development technics. The 
hoped-for result is that the cinematographer can devote more atten- 
tion to the artistic side of his work with a resulting increase in effec- 
tiveness of scene illumination. 



* Presented at the 1941 Fall Meeting at New York, N. Y. ; received October 20, 
1941. 

281 



282 STUDIO LIGHTING COMMITTEE REPORT [J. s. M. p. E. 

Increasing numbers of small single-lamp "broads" for use with 
750-w and 500-w "MP" and "CP" lamps are now making their ap- 
pearance on the sets. 

In Technicolor photography the use of "Inkie" light on the sets 
seems to be increasing. The use of the "CP" lamps with the Mac- 
beth Whiterlite filters in the equipment has proved highly successful. 
In the past year cameramen have been taking advantage of the flexi- 
bility of "Inkie" equipment and the large variety of available sizes 
and possibilities for placement and concealment that they present, to 
bring out to the fullest the great potentialities of beauty in color 
photography. 

A trend toward the use of "Inkie" equipment for top and key-lights 
in the illumination of small and medium-size sets is noted. The 5-kw 
and 2-kw "Solar spot" equipments are gaining in popularity, and 
large numbers of these comparatively small units are now employed. 

At the present time carbon arc lamps are being freely used for 
black-and-white as well as color photography. Each type of light- 
ing unit, whether carbon arc or incandescent tungsten, has its proper 
place in set lighting, and the choice of the correct equipment for the 
work at hand is a part of the cinematographer's individuality. 

As an example of placement of carbon arc lamps for black-and- 
white photography, the cinematographer on the picture Citizen Kane 
used Duarc broadside lamps as the major lighting units on many sets 
in that picture. 

Citizen Kane was a picture of extreme realism and characterization. 
Unlike most productions, the sets were ceilinged, which made it 
necessary to do most of the lighting from the floor. Such procedure 
required lamps with the penetrating power of the carbon arc. 

In obtaining unusually great depth of field, apertures of the order 
f //9 //H> an d even//16 were used. This called for lighting units 
with great illuminating power in comparatively small housings. 
Duarcs or type 170 carbon arc spotlights were used for this work. 

It is recognized that this is an unusual type of picture, and may not 
be indicative of any trend for future lighting. 



R. G. LINDERMAN, Chairman 

F. E. CARLSON C. W. HANDLE Y K. FREUND 

R. E. FARNHAM D. B. JOY K. STRUSS 

J. W. BOYLE 



Mar., 1942] STUDIO LIGHTING COMMITTEE REPORT 283 

DISCUSSION 

MR. OFFENHAUSER: What are the name and the type number of the filter 
used to correct the color- temperature of the lamp referred to in the report? 

MR/LINDERMAN:* The filter is the Macbeth Whiterlite filter. It is usually 
used in the shape of a rondel. Three sizes are furnished, which fit inside the 
500-watt, 2000-watt, and 5000-watt Solarspots. 

MR. OFFENHAUSER : What is the color-temperature of the lamp used, and to 
what color-temperature is the source raised through the use of the filter? 

MR. LINDERMAN:* The color-temperature of the CP (color photography) 
lens used is rated at 3380 K. 

The exposure meter mentioned in the report is the one developed by Captain 
Don Norwood, described in a paper presented at the 1940 Fall Meeting at Holly- 
wood and published in the April, 1941, JOURNAL (p. 389). 



Communicated. 



ADVENTURES OF A FILM LIBRARY* 
RICHARD GRIFFITH** 



Summary. Collecting and circulating important films of the past is not as dusty 
an occupation as it sounds. Even the mechanical acts of collecting and preserving film 
involve the human factors: people feel strongly about works that they themselves have 
created, criticized, or merely seen, and the collection of films both in this country and 
in Europe has been fraught with emotional, financial, and political complications, 
while the number of illustrious personalities who have in one way or another become 
involved in the Museum of Modern Art Film Library's work is evidence of the ability 
of even the most ancient fragments of celluloid to retain a contemporary as well as an 
archaeological interest. 

Circulation of the Film Library's motion picture programs has proved illuminating 
in its revelation of the attitude taken toward the film medium by all varieties of persons. 
The purpose has been to provide students with the opportunity to form a critical atti- 
tude by examining important films at first hand. There has gradually grown up a new 
appreciation which has learned not only to marvel at the rapid development of this new 
medium but also to discern its enormous and largely untapped potentialities. 

To "trace, catalog, assemble, exhibit, and circulate a library of 
film programs" the stated purpose of the Museum of Modern Art 
Film Library when it was founded in 1935 sounds like a straight- 
forward enough job. A difficult one, perhaps, and a dead and dusty 
one to some, though not to the trustees of the Museum of Modern 
Art and the staff of the infant Film Library, for this department of 
the Museum was created, in the words of the original announcement, 
"so that the motibn picture may be studied and enjoyed as any other 
one of the arts is studied and enjoyed." Everyone connected with 
the Museum and the Film Library expected to enjoy this work of re- 
viving favorite films of the past as much as did the groups of movie en- 
thusiasts throughout the country who have since constituted the 
Film Library's audience. "Tracing, cataloging, and exhibiting" im- 
portant pictures seemed about the pleasantest or one might even 
say the most glamorous form of scholarship imaginable, and no one 

* Presented at the 1941 Fall Meeting at New York, N. Y.; received October 
3, 1941; adapted from an article by Iris Barry, Curator of the Film Library of 
the Museum of Modern Art. 

** Museum of Modern Art, New York, N. Y. 
284 



ADVENTURES OF A FILM LIBRARY 285 

supposed, at that time, that it could present problems much beyond 
those to be expected from any scholarly activity. 

That was six years ago. Since then the Film Library, with the 
aid of a grant from the Rockefeller Foundation and gifts of money 
from private sources, has accumulated sixteen million feet of film: 
it would take 3000 hours, or 365 eight-hour days, to see it all. During 
that same period the Library has circulated ninety-one programs of 
films selected from this collection to 476 museums, colleges, and other 
educational or non- commercial institutions, some of which have taken 
virtually everything that the Film Library was able to supply, while 
others have several times repeated series of programs. The films cir- 
culated are all newly made positive prints and the majority of them 
go out with printed program notes of a critical and descriptive na- 
ture for each member of the audience, together with piano music to 
supply an accompaniment for silent films. Book lists, bibliographies, 
and information of a scholarly or technical nature have been freely 
provided not only to users of programs, to journalists and writers, 
to firms or individuals in the motion picture industry, but to the pub- 
lic generally. The Film Library has in fact become a center of in- 
formation about all matters pertaining to films. At the same time 
it has amassed a comprehensive collection of books, periodicals, 
manuscripts, still photographs, and other materials on the motion 
picture which is now available for reference without charge to the 
general public through the Museum Library. 

Statistically speaking, this sounds as if the original purpose of the 
Film Library has been carried through as conceived as a plain and 
straightforward museum job. But the statistics fail lamentably to 
tell the story. The Film Library soon discovered in fact what it had 
all along believed in theory, that the movie is the liveliest of the arts, 
and that even the most ancient fragments of celluloid retain a con- 
temporary as well as an archaeological interest. Almost all the 
Library's activities have been fraught with emotional, financial, and 
even political complications, and the task of creating this film collec- 
tion has proved to be a long series of adventures, in which the human 
element has loomed larger than scholarship or finance. 

In 1935, no consistent effort had been made to preserve or present 
for reexamination the films of the past through which, step by step, 
this great and popular art has developed. Almost all the films of 
historical or aesthetic importance were then invisible and in danger 
of being lost or destroyed. Under such circumstances, no serious 



286 R. GRIFFITH [j. s. M. P. E. 

study of the history and development of the motion picture or of its 
influence or aesthetic value could be undertaken. The Film Library, 
therefore, conceived as its first and most immediate task the collec- 
tion and exhibition of the maximum amount of film. 

Here it was, of course, a case of first catch your hare. How were 
the necessary films to be obtained? It is not widely realized that a 
motion picture can not usually be bought or otherwise procured as 
can a book or a painting; or that, even if a print of a film be so ob- 
tained, its physical possession does not necessarily entail the right to 
its use or showing. It is true that in the early days of film history 
there were many producers, like the Frenchman Georges Melies, who 
sold prints of their films outright, so that a purchaser could dispose 
of a copy of, say, A Trip to the Moon, exactly as he wished until it 
wore out. Here were obvious disadvantages. The producer could 
flood the market with as many prints as he could sell to competing 
showmen, yet it was generally they who reaped the profits through 
exhibition, and there was no way of ensuring that the purchaser of a 
print would not, as he often did, illegally make duplicate negatives 
from his print and so compete with the producer himself in the sale 
of still more prints. It is also true that in those early days a success- 
ful film continued to be shown for a long period of time (for all we 
know, A Trip to the Moon is even now being shown commercially 
in Zanzibar or in the Australian interior) instead of, as today, for a 
mere year or so. But most of the men who made the early films went 
bankrupt. From a collector's point of view it is fortunate that their 
methods of business left behind a residue in the form of numbers of 
prints in private hands or in the vaults of dealers in scrap-film. And 
so, anomalously, the Film Library often found it easier to acquire 
very old films than more recent ones. 

It proved, in fact, relatively simple to acquire by token purchase, 
gift, or loan a fair representation of film-making in this country and 
abroad from 1895 to 1912. Prints of primitives of the art like The 
Great Train Robbery were acquired from the widow of the pioneer 
Jean A. Leroy. From amateurs, from scrap-film or "junk" firms, 
from film pioneers or their heirs most of the outstanding early films 
were obtained. In one instance Mr. William Jamison of the Film 
Library's staff found a print of The Execution of Mary Queen of Scots 
(much spotted with tobacco juice) in an open garbage can in the 
Bronx. Another day a total stranger telephoned to offer a film that 
she had had in her hat-closet for many years. 



Mar., 1942] ADVENTURES OF A FlLM LIBRARY 287 

The situation proved quite otherwise in regard to films of later 
date. Most of the motion pictures made since 1912-1914 are the 
property of producer or producer-distributor firms who rent but do 
not sell prints for commercial exhibition through their own or other 
distributing companies. Ownership and consequently the right to 
exhibit such films remains firmly in these hands. Obviously, then, 
in order to gain access to such material, it was immediately neces- 
sary to enlist the sympathetic support of the film industry as a whole. 
This the Film Library consequently attempted to do. Happily its 
creation, and the fact of its support by such an institution as the 
Rockefeller Foundation, had received a "good press"; people gen- 
erally approved the idea. And, equally happily, among the trustees 
and friends of the Museum were several who had immediate interests 
of one sort or another in the motion picture industry. 

Armed therefore with auspicious introductions, the director, John 
E. Abbott, and the curator, Iris Barry (in private life, Mrs. Abbott) 
shortly after the formation of the Film Library found themselves, to 
their considerable amazement, in Hollywood about to entertain an 
alarmingly brilliant concourse of filmdom's great at Pickfair. Cer- 
tainly the gracious gesture of Mary Pickford in thus throwing open 
her famous house, and herself acting as hostess, could alone have af- 
forded the infant Film Library so excellent an opportunity of putting 
its case before the aggregate film chiefs, and through the press, 
which reported the somewhat unusual event, a wider circle of film 
employees and the public. That evening, pioneers of the industry 
like Mack Sennett met newcomers like Walt Disney for the first 
time, old acquaintances were renewed and new ones made, for Holly- 
wood, contrary to report, is no more gregarious than any other com- 
munity whose inhabitants work for a living and reside, often, twenty 
miles apart. For once the exponents of this new art-industry who 
normally live for the immediate future and the work in hand, stopped 
the clock briefly to consider the past. They had been, for the most 
part, brilliantly successful but had often as a consequence faced 
criticism proportionate to the enormous influence they exercised on 
the public imagination. Now an outside agency suddenly appeared, 
asking only permission to preserve a record of their endeavors, and 
seeking no financial contribution from them at least not for the 
moment. The brief program of films given after supper had been 
selected by the Film Library as an illustration of the work it intended. 
This glimpse of the birth and growth of an art that was peculiarly 



288 R. GRIFFITH [J. S. M. P. E. 

their own both surprised and moved this unique audience. The 
screen doubtless brought back memories of early struggles and half- 
forgotten triumphs, of former companions seldom remembered. 
There was a tiny, shocked gasp at the first appearance of Louis 
Wolheim in the program's brief excerpt from All Quiet on the Western 
Front; he had been dead so very short a time. Was fame so brief? 
Of course there ought to be a museum of the film! At the close of the 
program, Will H. Hays and Mary Pickford endorsed the Film Li- 
brary's undertaking in enthusiastic speeches. Samuel Goldwyn 
and Harold Lloyd as well as Miss Pickford promised their films. A 
major obstacle had been overcome in gaining the attention and un- 
derstanding of the industry as a whole. It remained to work out a 
basis on which films owned by large corporations rather than by in- 
dividual producers could be made available to the Film Library. Its 
principles had been accepted even if there remained individuals to 
convert, like the world-famous director who, that same summer of 
1935, said amiably but firmly that he, for one, was not interested in 
the preservation of his own films and that nothing could convince 
him that films have anything to do with art. It is pleasant to record 
that he has latterly become a warm supporter of the Film Library's 
undertakings. 

Back in New York, a satisfactory basis of operations was reached 
after considerable negotiation with the big producer-distributor com- 
panies involved, and the Film Library finally obtained the right to 
the non-commercial use of industry-owned films. This was a signal 
triumph, for not only could the institutions obtain and use prints of 
the films it wanted but the propriety of its undertaking had thus been 
recognized. There were stipulations that bore heavily on some of 
the institutions desirous of using the programs, but some of these, too, 
are about to be relaxed, and other problems that arose pertaining to 
copyrights and the like have been solved as they came up. 

There is much misconception regarding the nature of film and its 
use; the general public is not wholly aware that it is a costly com- 
modity to handle. It is often asked why the Film Library can not 
purchase "old" film outright. Other considerations aside, this is 
impossible since the copyright title of almost any film can be resold 
for, and consequently is worth, many thousands of dollars. The 
big producer-distributor firms preserve all their negatives but, us- 
ually, not prints. Were they to preserve prints that have seen much 
use, or preserve new prints over a long period of time, they would be 



Mar., 1942] ADVENTURES OF A FILM LIBRARY 289 

useless for projection in any case. In order to obtain any picture, 
the Film Library consequently pays the actual cost of making up 
the new print or prints it requires and of replacing these prints when 
they wear out. It should be borne in mind that no non-commercial 
institution could afford to accept a print of each film made, even at 
no charge as the Library of Congress acquires copies of all newly 
published books because the cost of storage and preservation would 
be prohibitive. The Film Library can only select. 

Here arises a question : what shall be selected either for preserva- 
tion and research or for circulation to the wider circle of film students? 
Little doubt arises in the case of greatly famous films like The Birth of 
a Nation, The Cabinet of Dr. Caligari, The Four Horsemen of the Apoca- 
lypse, Potemkin, The Jazz Singer. Though these may seem quaintly 
archaic or regrettably controversial, nevertheless they are "musts." 
But what of others that, because of their extreme popularity at one 
time, or because they incorporated a technical or aesthetic innova- 
tion, or celebrated a social or political trend, might be needed in a 
well rounded collection. Take the case of Will Rogers' The Headless 
Horseman, first feature-film on panchromatic stock. Or the brief 
Peace of Britain, which urged the case against war. Or The Andalu- 
sian Dog, without which no study of surrealism or the work of Salva- 
dor Dali would be complete. It would be difficult to view the Rogers 
film without boredom, while some proportion of any general audi- 
ence would doubtless feel indignation upon viewing The Peace of 
Britain and considerable shock upon seeing The Andalusian Dog. 
Boring or tendentious books may go on a library shelf; paintings 
that provoke strong reactions may be hung publicly without exces- 
sive odium. The exhibition of a film has not yet become so liberal 
an undertaking. Individuals violently protest against the spectacle 
of films that for any reason displease them: custom seems to have 
entailed upon the showing of films a peculiar moral responsibility. 
The seeing of films is of course a mass experience, not a private one 
like the reading of a book or the viewing of a painting, which may 
explain much. And perhaps it is felt that with this new medium of 
expression, as with the printed word long years ago when it too first 
arrived to extend the public experience, there is some special if mys- 
terious danger. After all, films do convey ideas! 

This is by no means the only problem that the collection and re- 
vival of films creates. Suppose there is a widespread opinion backed 
up by recollection and hearsay or by written criticism that a film such 



290 R. GRIFFITH [J. S. M. p. E. 

as Lubitsch's The Marriage Circle should definitely be included and 
shown. To the layman the solution might seem easy: one has only 
to look and see whether it survives the test of time. But before a 
film of that age can be looked at, a print must be made up and paid 
for and a $250 look proves costly indeed should the film turn out to 
be of restricted interest. More than that, it will cost $10 a year 
thereafter to keep the print in storage if one elects to wait and see 
whether, with the passage of still more time, the change of taste and 
outlook may show this judgment a misguided one as in the other 
arts many such judgments have been. In truth there can be few 
materials subject to scholarship and preservation that present more 
difficulties than do motion pictures. A mischievous muse presided 
at their birth. 

So in gathering together important but long unseen films of the 
past the Film Library staff, somewhat like Whistler, has had to rely 
"on thp experience of a lifetime." Its own aggregate experience in- 
cludes that of its dean, William Jamison (who has been in the indus- 
try since its birth and is rumored to be able to tell who made a film 
merely by feeling it between his fingers in the dark) and of associates 
like the ace cameraman "Billy" Bitzer, or particular friends in and 
outside the industry like Terry Ramsaye, Louis Bonn, or W. S. Van 
Dyke. There is also the valuable consensus of the public to draw 
upon : everyone agrees that the first Mickey Mouse and // Happened 
One Night are proper museum pieces, and everyone wonders why on 
earth the Film Library does not have Quo Vadis and The Blue Angel. 

First catch your hare! When two comprehensive series of pro- 
grams tracing the development of the American film were already in 
circulation, the director and the curator undertook a protracted trip 
in the summer of 1936 to London and Paris and thence to Berlin, 
Warsaw, Moscow, Helsingfors, Stockholm. The results were fruit- 
ful and sometimes quite unexpected. To begin with, the idea of 
collecting and preserving film was already not novel in Europe. 
Film archives of very different kinds but nevertheless archives 
had already been started in Paris (privately or at best with semi- 
official blessing and virtually no funds), in London (semi-officially 
and with some but inadequate funds for preservation as opposed to 
collection), in Berlin (officially and with apparently ample funds, 
with emphasis on preservation but still more on the collection of 
propaganda films, preferably anti-German), in Moscow (officially 
but apparently with little regard for preservation), and in Stockholm 



Mar., 1942] ADVENTURES OF A FILM LIBRARY 291 

(under ideal conditions but for Swedish films only). The emissaries 
of the Film Library were eagerly shown vaults of film, asked whether 
some exchange basis might be arrived at, or whether, if the threat of 
war came closer, New York might undertake to store precious 
prints or negatives. Paris was particularly sensitive to war risk: 
between 1914 and 1918 far too much historic French film had been 
junked to provide the nitroglycerine to which celluloid is chemically 
akin. The warmest understanding was found in France, and much 
film was acquired. In Berlin it was Olympics year and tourists were 
particularly welcome. It was a relief to find that, contrary to ru- 
mor, the films of Germany's great silent period 1919-1928 had not 
been destroyed, so that the Film Library was able to obtain virtually 
everything it desired, from The Cabinet of Dr. Caligari down to a 
brutal but most illuminating Nazi-inspired feature of 1934. More- 
over, from the official Reichsfilmarchiv came Potemkin, most famous 
of Russian propaganda films, the original negative of which Dr. 
Goebbels had purchased in 1933, pointing to it as a model for Nazi 
propagandists to emulate. 

For some reason that the director and curator never quite fath- 
omed, the sailing was less smooth in Moscow. It may merely have 
been that they represented an institution supported by private con- 
tributions and not by government funds. Certainly film directors 
and students were friendly and eagerly interested, but officialdom, 
through whom alone any film could be obtained, at first refused all 
occasions for presenting any request. Finally, with the strenuous 
help of the U. S. embassy, a hearing was obtained and the reluctant 
or suspicious atmosphere was somewhat overcome. A basis for ex- 
change of film between the Scientific Research Institute's collection 
was finally arrived at, and late in 1936 four films finally arrived in 
New York from the U.S.S. R. But none really representative of the 
new regime arrived until the late summer of 1939 after little hope re- 
mained of receiving even an acknowledgment of the films the Library 
had dispatched to the Institute. When they did arrive it was at a 
moment when, to put it mildly, recent political events had created a 
marked apathy toward Russian films, no matter how interesting they 
might be from a technical or ideological viewpoint. Nor was there 
much but apathy or distaste for films of comparable nature from con- 
temporary Germany, such as the propagandist documentary of 1937, 
Flieger, Funker, Kanoniere, in which Field Marshal Goering in per- 
son commended the effort that gave Germany an air force, while the 



292 R. GRIFFITH [J. S. M. P. E. 

film itself showed a glimpse of dive-bombing and introduced the now- 
familiar term Luftwaffe. It was nevertheless a shadowy image of 
what was to come both in fact and in later documentaries like Feldzug 
in Polen and Sieg im Westen. Recent events have led to a wide 
recognition of the use and value of studying propaganda material. 
But at that time the acquisition of foreign material of this kind gave 
rise to a whispering campaign (originating, it seemed, among small 
groups of film enthusiasts with axes to grind) that the Film Library, 
or the Museum as a whole, or perhaps even the Board of Trustees ( !) 
was infiltrated with Nazi principles (this was in 1937 and 1938) or 
with Communist principles (this was in 1940) or at best with some 
"un-American" spirit. 

Meanwhile the Film Library has restored to view, to admiration, 
and to their proper standing among other contemporary works of 
art, a great wealth of American films. It pointed out (what is per- 
haps obvious) that while France at first, then Italy, then Germany 
contributed both styles and film-makers to the growing American 
film industry, the motion picture as such is triumphantly and pre- 
dominantly an American expression. It was this American develop- 
ment, the Film Library stated, that in turn signally stimulated film- 
making in France, in the U.S.S.R., and indeed the world over by its 
technical inventiveness, its energetic and impulsive grasp of the 
medium. If the little campaign of slander persisted longer, it re- 
ceived small credence. Meanwhile in 1938, when at the invitation 
of the French government the Museum of Modern Art put on ex- 
hibition "Three Centuries of Art in the United States" at the Musee 
du Jeu de Paume in Paris, the section prepared by the Film Library 
was the most widely attended and, with the architectural section, 
the most enthusiastically received of all the comprehensive show. 
In films the United States was seen at its most original, most exuber- 
ant, most enjoyable, most understandable. And in presenting its 
"Brief History of the American Film" there, as in all its numerous 
programs circulated at home, the Film Library affirmed its unwaver- 
ing faith in the film as the liveliest as well as the most popular of the 
contemporary arts and one in which the United States is supreme. 

Public education and guidance in film appreciation have been so 
slow to develop, however, that people sometimes complain that they 
do not "like" all the films shown, forgetting that these are not shown 
as diversion or entertainment but for the pleasures of comparison, 
analysis, and study. A few make a small nuisance of themselves by 



Mar., 1942] ADVENTURES OF A FlLM LIBRARY 293 

rather ostentatiously tittering at the outmoded dresses, obsolete 
slang, old-fashioned moral values of films ten or twenty years ago. 
This, it must be said, is habit fostered by certain sections of the film 
industry itself through the revamping of "old" films to turn them 
into ridicule. But it is interesting to observe that films which are 
old enough do not provoke that reaction. It is very evident, too, 
that laughter at the death of Camille, played most expertly though 
in an obsolete style by Sarah Bernhardt, or at the dresses of Greta 
Garbo in Susan Lennox, is fraught with shock at the sudden disrup- 
tion of the time sense rather than with merriment. As audiences 
gain the habit of looking at films as something more than a transient 
distraction, the tendency to ridicule diminishes noticeably, but its 
existence suggests some curious conclusions on the impermanence 
of standards of taste. 

Aware of all these and other difficulties and criticism it has faced 
and must expect, the Film Library has prepared for exhibition a very 
large proportion of its collection in the current retrospective of mo- 
tion picture history, "A Cycle of 300 Films, 1895-1940," which now is 
showing daily in the Museum's auditorium, and which contains the 
cream of the Library's collection so far. No such experience has 
ever been afforded to the students and enthusiasts in any art, let 
alone in the art of the motion picture. That it is possible at all after 
so few years of activity in this new field is due largely to the continued 
enthusiasm and support of the Museum trustees, of the representa- 
tives of the motion picture industry on its Advisory Board, and of 
those amateurs of cinematography who attend its showings. 

There are some notable absences in the cycle. Where, for instance, 
is Quo Vadis, epic costume film of 1911 from Italy, which played so 
signal a part in introducing longer pictures and in making the movies 
respectable? The most patient search and inquiry has brought no 
trace of it thus far. Morocco is here but not the well-remembered 
Blue Angel, in which husky- voiced Marlene Dietrich broke like a new 
comet on our world. Owned originally by one of the big American 
firms for distribution in this country, the rights had reverted to the 
German producer-firm, Ufa. This was one film asked for in Berlin 
in 1936 that Nazi officialdom refused even though, surprisingly, they 
authorized the acquisition of numerous other films made by non- 
Aryans and anti-Nazis, most of whom now swell Hollywood's rank 
of talent. What about The Exploits of Elaine, which "everybody" 
remembers as having been so incomparably breath-taking? This 



294 R. GRIFFITH 

was acquired, via Paris, from Brussels where someone had tracked 
down a series of episodes which were purchased for 1000 francs. 
The prints on arrival proved in such bad condition, so brittle and with 
so many sprocket-holes broken, that a duplicate negative of one reel 
was made up for tentative inspection before venturing to "dupe" 
the whole eight episodes. Never was there a sadder little group of 
film fans than the staff of the Film Library upon viewing that trial 
restoration of one of the great legendary "musts" of film history. 
Hopefully, another episode was printed : it seemed a little better, but 
unless further inspection of further episodes brings higher quality, 
then Elaine must join the few other overrated "gems of the past." 

Unfortunately lacking are many of the Mack Sennett comedies of 
the vintage period. Inquiry has proved that, not so very long ago, 
the original negatives of these films were cut up to be remade into 
talkies. All was lost in this case. Regrettably, the old John Bunny 
and Flora Finch comedies proved disappointing, rather flat and un- 
funny: perhaps others will be found that will revive memories of 
childhood enjoyment long years ago. 

And where, especially, are the Charlie Chaplins? True, there are 
a few here in the cycle, early and delicious, but so few. Others, 
slightly later and still better, which were shown earlier at the Mu- 
seum, are no longer visible. These unhappily must remain in the 
vaults, for they may no longer be shown at least for the time be- 
ing. As suggested earlier, the Film Library is the custodian, not the 
owner of its films, and the owners may require it to desist from ex- 
hibition at any time. This happens when pictures in the collection 
are revived for commercial exhibition. It seems hard, especially in 
cases where it is reasonable to suppose that interest in them was first 
recreated by the Film Library's revival of them. As for the later 
and greater Chaplin comedies, these belong to the comedian himself, 
and so far he has not seen his way to turning them over. 

Meantime, with exceptions as noted, the Film Library has con- 
tinued to fulfill its original purpose and has further extended its pro- 
gram. In its newest phase of activity, it houses a project under the 
Office of the Coordinator of Cultural and Commercial Relations be- 
tween the American Republics for the provision of educational and 
documentary films in Spanish and Portuguese to the twenty sister na- 
tions of the hemisphere. With the growing recognition of the film as 
a major social and cultural force in our time the Film Library's 
horizon becomes unlimited. 



WORK SIMPLIFICATION ESSENTIAL TO DEFENSE* 
ALLAN H. MOGENSEN** 

Summary. Increasing productivity is the most important job today. It will not 
be done by "speeding up." Work simplification, using motion pictures as the main 
tool, helps everyone in an organization to find ''the one best way." Applications of 
the motion picture in such studies are described in the paper, as also a specially de- 
signed projector for exhibiting the pictures. 

The most serious problem facing America today is that of increas- 
ing our industrial productivity. No matter how much money we 
spend, no matter how much floor space we add to our plants, no 
matter how many machines we eventually are able to build, we are 
nevertheless limited by the number of possible man-hours. And 
when one analyzes the figures of the new plans and schedules, it is 
immediately apparent that unless something is done to increase pro- 
ductivity we shall not be able to increase our production to the point 
where it will be of real value in winning the war. 

Few people understand the difference between increasing produc- 
tion and increasing productivity. So far we have gone largely on the 
assumption that if we want to turn out twice as much product, we 
merely have to double the floor-space, buy twice as many machines, 
and hire twice as many men. Temporarily the question as to where 
the money is coming from does not bother us, but it most certainly 
will if the effort is continued over a period of several years. Up to 
the present, floor-space has not been a serious handicap as one can 
readily see from the sizes of the aircraft and other plants that are 
being put up all over the country. However, we are now reaching 
the point where shortage of building material may be a critical fac- 
tor. Securing adequate productive machinery has been a problem 
all through the defense effort, and would be an increasingly serious 
one if the program is multiplied in its scope. 

Man-power, on the other hand, has been a serious problem from 

* Presented at the 1941 Fall Meeting at New York, N. Y.; received Novem- 
ber 1, 1941. 

** New York, N. Y. 

295 



296 A. H. MOGENSEN (J. S. M. P. E. 

the start, although taking up the slack of unemployment and then 
robbing skilled men from other "less important" industries has so 
far seen us through the preliminaries. It soon became evident, how- 
ever, that it would be necessary to train many of the unskilled people 
for skilled jobs, and to induct into unskilled jobs many who have 
never worked before. Witness the effort of many of the fine men 
who have been called from industry to come to Washington to assist 
in setting up training programs and in developing the "up-grading" 
plans that have produced results. 

All this, however, is overlooking the most obvious truth of all, and 
that is that if we are to increase output we must increase produc- 
tivity, that is, we must increase the efficiency of every unit in our 
production scheme. If a man can produce twice as much with the 
same equipment, floor-space, and in the same number of hours, and 
yet go home less tired than he was before, something has been ac- 
complished toward finding a solution to a problem that has puzzled 
so many people. Notice I say in the same number of hours. In the 
last war England tried very long working hours in an effort to increase 
output. Studies made by the National Institute of Industrial Psy- 
chology proved conclusively that output fell off rather than increased, 
and therefore the number of hours was reduced. Despite this lesson 
learned both abroad and here during the last war, England has again 
attempted, and we are now attempting, to increase our output by 
going to excessive overtime. 

What is the answer? Work simplification, taught to every single 
member of an organization and taught by the use of motion pictures, 
is the answer. 

Very few persons understand the difference between work done 
at high speed and work done in a hurry. Most of the attempts made 
so far at increasing our output have been to "speed up" production. 
Work done at high speed will give good results because it is accom- 
plished by eliminating unnecessary parts of the job; whereas work 
done in a hurry will be unsatisfactory because it includes speeding 
up both necessary and unnecessary parts of the job. One can not 
blame the workman for not understanding this fundamental fact 
when very few of our chief executives appreciate its truth. There- 
fore, instead of trying to get a man to work harder or faster, work 
simplification seeks to find the "one best way" of doing each job. 
This way is usually the easiest way and, in fact, is often so simple 
that one wonders "why we did not think of that before." 



Mar., 1942] WORK SIMPLIFICATION 297 

The motion picture is used in two ways in connection with work 
simplification programs. The first and most obvious form is the 
"before and after" picture to teach fundamentals to all concerned. 
Such films show the former way of doing a simple operation, and then 
demonstrate the new and easier way of performing it. Each of the 
films illustrates a fundamental principle in work simplification or 
motion study, and then enables the person who has seen the film to 
analyze his own operations and apply the principle to simplifying 
them. 

(Mr. Mogensen then showed a film on a simple operation in a garment plant 
demonstrating how the output of a unit was tripled by analyzing the motions 
that went into the performance of this operation. Part of the increase was due 
to the elimination of seventy-four useless motions formerly included, and part 
was due to the fact that the new method was so much simpler that it was possible 
for the girls to develop much higher skill than previously. Also, many more 
girls could become really skilled operators under the new method, whereas only a 
few could become skillful under the previous method. Redesign of the work- 
place, study of the tools used, and operator training were demonstrated in the 
picture.) 

The use of the motion picture in this way is, of course, not new. 
One application that has been extremely successful, especially in 
some of the new defense plants, is the use of a film-loop for training 
operators. The idea of the film-loop used for training is not new. 
Frank B. Gilbreth applied it at least twenty years ago, and the 
speaker has used it in connection with training programs since 1926. 
However, proper equipment for the most effective use of this tool 
has not been available until recently. A special version of a com- 
mercial 16-mm projector has been designed by Prof. David B. Por- 
ter, of New York University, and has been used in almost every con- 
ceivable kind of industry and business by graduates of the work 
simplification conferences held each year at Lake Placid Club. 

The use of the equipment can be illustrated by describing the pro- 
cedure used in training operators in a new defense plant recently set 
up in the Middle West. First, the operation as it was performed at 
one of the U. S. arsenals was photographed completely from start to 
finish on 16-mm film. The picture ran about 4000 feet when edited 
and titled. A small group of engineers then looked at the film, and 
of each operation asked, "Why?" "Is there a better way?" 

Analysis of the film occupied the time of the new plant manager 
and an industrial engineer for about a week; at the end of this time 
it became evident that the specifications set forth by the government 



298 A. H. MOGENSEN [j. s. M. P. E. 

demanded about twice the equipment that was actually necessary. 
It was shown that the new plant could be erected with about half the 
floor-space called for in the government specifications, and that only 
about half the number of operators would be required. It can be 
seen, of course, that this greatly simplified the problem of procuring 
machinery, as well as that of training operators. 

As each new employee was brought into the plant, he or she was 
shown the picture in its form at that moment and told that the pro- 
cedure depicted was not necessarily the best way to make the prod- 
uct, but that it was the best way known at that time. Each em- 
ployee was told that he or she would undoubtedly have many good 
suggestions for simplifying the operations and were urged to ask 
"Why?" at every step. Aptitude tests, followed by a medical ex- 
amination, then determined whether or not the applicant was suit- 
able for the job for which he or she was being considered. 

When the operator was finally selected for the job, he (or she) was 
taken into the training room and the particular loop of film covering 
the operation was projected on the screen. The film was run slowly 
at first, and all the requirements as to quality of product, safety, and 
output were explained. Usually several loops covered an operation. 
First, the new employee would be shown a long shot demonstrating 
the overall handling of the equipment, the procurement of the ma- 
terial, and general safety factors surrounding the operation. Then 
successive close-ups were shown, and every conceivable feature of 
the operation was demonstrated and explained by the instructor. 

Following the instruction with the film-loop, the operator was in- 
troduced to the foreman or forelady on the job and the actual train- 
ing was begun. Successive visits to the classroom were included as 
the training progressed so that the correct procedures could be con- 
stantly emphasized. Very often in the past instruction cards or 
write-ups of the correct methods have been given to new operators, 
but no attempt has been made to see that the operators actually 
followed the designated motions during the training period. Bad 
habits creep in, and while often they are small and may appear in- 
significant, they spell the whole difference between a smooth, effi- 
cient, and restful operation as against the nervous drive so often 
seen behind attempts to get operators to increase their output. 

One of the very important advantages of the special projector is 
that one does not have to take all the pictures at 64 frames a second 
in order to slow down the motion. The biggest objection to slow 



Mar., 1942] WORK SIMPLIFICATION 299 

motion in the past has been the expense of filming everything at 
that speed. Very often only a small part of the resulting film was 
needed in slow motion and the cost was therefore prohibitive. With 
this machine everything can be taken at 16 frames a second, and the 
slow-motion effect secured in almost all instances except where very 
rapid hand motions do cause some blur when the still picture is pro- 
jected on the screen. 

For actual motion analysis two new features have greatly simplified 
the determination of time from the film. In the days of Gilbreth's 
first work the timing device was a spring-driven "microchronome- 
ter," later replaced by a Telechron motor. The latest develop- 
ment is a small device that eliminates the necessity of reading a 
dial inasmuch as it is built on the cyclometer principle. It is pro- 
duced by Veeder-Root, Inc., and is known as the "Wink Counter." 

Where it is not desirable to include the clock in the film, one may 
take the picture with a constant-speed camera driven by a synchro- 
nous motor, and then the special projector demonstrated here is 
equipped with a small frame-counter so that time may be taken di- 
rectly from the film. A heat-absorbing filter allows lengthy study 
of each frame, if desired, without warping the film. There are many 
16-mm projectors on the market that are satisfactory for home pro- 
jection of an occasional frame, but none of them allows detailed 
frame-by-frame analysis of the type desired by the motion study 
engineer. The 16-mm motion picture film provides the only effec- 
tive means of making micromotion studies where high production 
work warrants the cost of such analysis. 

Motion picture films are absolutely essential if the principles of 
work simplification are to be applied to the job of increasing produc- 
tivity. First of all, it trains the organization in finding easier and 
better ways of doing their jobs. It helps to break down the tradi- 
tions that so often dictate methods that are used largely because 
"we have always done it that way." The film loop has proved its 
value in enough instances to convince those who have used it that 
training time can be greatly reduced. An operator who is not suited 
for a particular job can be transferred to other work before a long 
period of discouragement and dissatisfaction is encountered. Super- 
visory and executive time is definitely reduced. Operators rapidly 
attain the required skill and thus fall into the high earning classes. 
Finally, this method of training will eliminate the need for much of the 
expensive equipment that was formerly tied up in the "vestibule 
school" method of training. 



CURRENT LITERATURE OF INTEREST TO THE MOTION PICTURE 

ENGINEER 



The editors present for convenient reference a list of articles dealing with subjects 
cognate to motion picture engineering published in a number of selected journals. 
Photostatic copies may be obtained from the Library of Congress, Washington, D. C., 
or from the New York Public Library, New York, N. Y. Micro copies of articles 
in magazines that are available may be obtained from the Bibliofilm Service, Depart- 
ment of Agriculture, Washington, D. C., at prevailing rates. 



Acoustical Society of America, Journal 

13 (January, 1942), No. 3 

Performance of Broadcast Studios Designed with Con- 
vex Surfaces of Plywood (pp. 244-247) 

Acoustic Impedance of Porous Materials (pp. 248-260) 
An Acoustic Tube for Measuring the Sound Absorption 
Coefficients of Small Samples (pp. 261-264) 

Properties of the Dulled Lacquer Cutting Stylus 

(pp. 265-273) 
A Large Radius Stylus for the Reproduction of Lateral 

Cut Phonograph Records (pp. 274-275) 
Tracing Distortion in the Reproduction of Constant 

Amplitude Recordings (pp. 276-280) 
A Noise and Wear Reducing Phonograph Reproducer 

with ControUed Response (pp. 281-283) 
The Correlation between Elastic Deformation and 

Vertical Forces in Lateral Recording (pp. 284-287) 

The Recording Laboratory in the Library of Congress 
(pp. 288-293) 

Educational Screen 

21 (January, 1942), No. 1 

Motion Pictures Not for Theaters (pp. 14-17, 21), 
Pt. 33 

Electronic Engineering 

14 (January, 1942), No. 167 
Television Picture Storage (pp. 578-580, 600) 
Frequency Modulation (pp. 584-585, 600), Pt. Ill 

300 



C. P. BONER 
L. L. BERANEK 

D. P. LOYE AND 
R. L. MORGAN 

C. J. LEBEL 
J. D. REID 
L. W. SEPMEYER 
F. H. GOLDSMITH 

S. J. BEGUN AND 
T. E. LYNCH 

J. B. WIESNER 



A. E. KROWS 



A. H. ROSENTHAL 

K. R. STURLEY 



CURRENT LITERATURE 



301 



Secondary Electron Problems in Beam Tetrodes 
(pp. 586-587) 

Electronics 

15 (January, 1942), No. 1 
Super-Cardioid Directional Microphone (pp. 31-33, 

91-93) 

International Photographer 

14 (February, 1942), No. 1 

Ninety Below (pp. 3^1) 

Government Training Films (pp. 8, 16) 

International Projectionist 

16 (November, 1941), No. 11 

Some Improved Methods of Controlling Carbon Arc 
Position (pp. 9-10, 12-13, 23) 



Projection Room Routine under Normal and Emer- 
gency Operation (pp. 15-16) 

New Photographic Lens Wins Trade Press Approval 
But I. P. Dissents (p. 19) 

RCA Review 

6 (January, 1942), No. 3 
NBC Studios 6A and 6B (pp. 259-268) 
A New Chemical Method of Reducing the Reflectance 
of Glass (pp. 287-301) 



J. H. OWEN HARRIES 



B. B. BAUER 



D. J. ZAFFARANO, 
W. W. LOZIER, 
AND D. B. JOY 

H. RUBIN 



G. M. NIXON 



F. H. NICOLL 



FIFTY-FIRST SEMI-ANNUAL CONVENTION 

OF THE 
SOCIETY OF MOTION PICTURE ENGINEERS 



HOLLYWOOD-ROOSEVELT HOTEL 

HOLLYWOOD, CALIF. 
MAY 4th-8th, INCLUSIVE 



OFFICERS AND COMMITTEES IN CHARGE 

EMERY HUSE, President 

E. A. WILLIFORD, Past-President 

H. GRIFFIN, Executive Vice-President 

W. C. KUNZMANN, Convention Vice-President 

A. C. DOWNES, Editorial Vice-President 

J. G. FRAYNE, Chairman, Pacific Coast Section 

C. W. HANDLEY, Chairman, Local Arrangements Committee 

S. HARRIS, Chairman, Papers Committee 



G. A. CHAMBERS 
C. R. DAILY 



Pacific Coast Papers Committee 

R. R. SCOVILLE, Chairman 
F. L. EICH 
W. W. LINDSAY, JR. 



S. P. SOLOW 
W. V. WOLFE 



Reception and Local Arrangements 



J. O. AALBERG 
B. B. BROWN 
G. A. CHAMBERS 
W. E. GARITY 

A. M. GUNDELFINGER 

E. H. HANSEN 

J. K. HlLLIARD 
E. M. HONAN 



C. W. HANDLEY, Chairman 

B. KREUZER 

R. G. LlNDERMAN 

C. L. LOOTENS 

R. H. McCULLOUGH 

W. C. MILLER - 
G. S. MITCHELL 
K. F. MORGAN 
H. MOYSE 



W. A. MUELLER 
G. F. RACKETT 
H. W. REMERSHIED 
ALSTON RODGERS 
L. L. RYDER 
S. P. SOLOW 
H. G. TASKER 
J. R. WILKINSON 



F. ALBIN 
L. W. CHASE 

302 



Registration and Information 

W. C. KUNZMANN, Chairman 
J. FRANK, JR. 
J. G. FRAYNE 
C. W. HANDLEY 



S. HARRIS 
F. L. HOPPER 



L. A. AlCHOLTZ 

J. W. BOYLE 
J. L. COURCIER 



1942 SPRING CONVENTION 

Publicity 

JULIUS HABER, Chairman 
G. R. GIROUX, West Coast Chairman 
S. HARRIS 
S. E. HAWKINS 



303 



G. S. MITCHELL 
E. C. RICHARDSON 
R. R. SCOVILLE 



Luncheon and Banquet Committee 

L. L. RYDER, Chairman 
]. O. AALBERG EMERY HUSE 

J. G. FRAYNE H. T. KALMUS 

C. W. HANDLE Y M. S. LESHING 

E. M. HONAN N. LEVINSON 



R. H. McCULLOUGH 

W. C. MILLER 

P. MOLE 

H. G. TASKER 



A. C. BLANEY 
D. J. BLOOMBERG 
L. F. BROWN 
J. P. CORCORAN 



J. O. AALBERG 

J. DURST 

G. M. FARLY 

B. FREERICKS 

W. E. GEBHART, JR. 



Hotel and Transportation 

G. A. CHAMBERS, Chairman 
C. R. DAILY 
C. DUNNING 
W. C. HARCUS 
G. T. LORANCE 

Convention Projection 

C. L. RUSSELL, Chairman 
L. D. GRIGNON 
J. K. HILLIARD 
A. E. JACKSON 
W. W. LINDSAY, JR. 
R. H. MCCULLOUGH 



H. R. LUBCKE 

F. O'GRADY 

J. W. STAFFORD 
W. L. THAYER 



S. M. PARISEAU 
H. W. REMERSHIED 
C. R. SAWYER 
G. E. SAWYER 
H. A. STARKE 



Officers and Members of Los Angeles Projectionists Local No. 150 



Ladies' Reception Committee 

MRS. EMERY HUSE and MRS. J. G. FRAYNE, Hostesses 



Assisted by 



MRS. G. A. CHAMBERS 
MRS. F. L. EICH 

MRS. A. M. GUNDELFINGER 

MRS. C. W. HANDLEY 
MRS. J. K. HILLIARD 
MRS. E. M. HONAN 
MRS. B. KREUZER 
MRS. N. LEVINSON 
MRS. R. H. MCCULLOUGH 
MRS. G. S. MITCHELL 

MRS. W. V. 



MRS. P. MOLE 
MRS. K. F. MORGAN 
MRS. W. A. MUELLER 
MRS. G. F. RACKETT 
MRS. H. W. REMERSHIED 
MRS. E. C. RICHARDSON 
MRS. L. L. RYDER 
MRS. R. R. SCOVILLE 
MRS. S. P. SOLOW 
MRS. J. R. WILKINSON 
WOLFE 



304 1942 SPRING CONVENTION [j. s. M. p. E. 

Color Print Exhibit Committee 

O. O. CECCARINI, Chairman 

L. E. CLARK C. DUNNING L. D. GRIGNON 

T. B. CUNNINGHAM R. M. EVANS A. M. GUNDELFINGER 

TENTATIVE PROGRAM 
MONDAY, MAY 4, 1942 

9:00 a.m. Hotel Lobby; Registration 
12:30 p.m. Terrace Room; Informal Get-Together Luncheon 

Addresses by prominent Hollywood members of the motion picture 
industry; names to be announced later 
2:00 Blossom Room; General Session 

8:00 Blossom Room; Technical Session 

TUESDAY, MAY 5, 1942 

9:30 a.m. Hotel Lobby; Registration 

This morning will be left open for a possible trip or other activity to 
be announced later 

2:00 p.m. Blossom Room; Technical Session 
8:00 Blossom Room; Technical Session 

WEDNESDAY, MAY 6, 1942 

9:30 a.m. Hotel Lobby; Registration 
10:00 Blossom Room; Technical Session 

2:00 p.m. The afternoon will be left open for a possible trip, to be announced 

later, or for recreation 
8:30 Blossom Room; Fifty-First Semi-Annual Banquet and Dance; details 

to be announced later 

THURSDAY, MAY 7, 1942 

10:30 a.m. Hotel Lobby; Registration 

Open morning 

2:00 p.m. Blossom Room; Technical Session 
8:00 Blossom Room; Technical Session 

FRIDAY, MAY 8, 1942 

10:00 a.m. Blossom Room; Technical Session 
2:00 p.m. Blossom Room; Technical Session 
8:00 Blossom Room; Technical Session 

Adjournment of the Convention 



Mar., 1942] 1942 SPRING CONVENTION 305 

HEADQUARTERS 

The Convention headquarters will be at the Hollywood-Roosevelt Hotel. 
Excellent accommodations have been assured by the hotel management at the 
following per diem rates: 

One person, room and bath $3 . 50 
Two persons, double bed and bath 5 . 00 

Two persons, twin beds and bath 6 . 00 

Parlor suite and bath, one person 8 . 00-14 . 00 

Parlor suite and bath, two persons 12 . 00-16 . 00 

Room reservation cards will be mailed to the membership early in April and 
should be returned to the hotel immediately to be assured of satisfactory ac- 
commodations . 

Indoor and outdoor parking facilities adjacent to the hotel will be available for 
those who motor to the Convention. 

Golfing privileges may be arranged by request of the hotel management or at 
the registration headquarters. 

Registration headquarters will be in the hotel lobby. All members and guests 
attending the Convention will be expected to register and receive their Conven- 
tion badges. The registration fees are used to defray the expenses of the Con- 
vention, and cooperation in this respect is requested. Identification cards will 
be supplied, which will serve as admittance to all scheduled or special sessions, 
studio visits, and trips, and several de luxe motion picture theaters on Hollywood 
Boulevard in the vicinity of the hotel. 

Members planning to attend the Convention should consult their local railroad 
passenger agents regarding train schedules, rates, and stop-over privileges en 
route. For a stop-over at San Francisco the Convention Committee recommends 
the Mark Hopkins Hotel, on "Nob Hill." Accommodations may be arranged 
with Mr. George D. Smith, manager of this hotel. 

An interesting color-print exhibit will be an adjunct to the Convention and will 
be open to the public and delegates during the five days of the Convention. 

The Convention hostesses promise an interesting program of entertainment for 
the visiting ladies. A reception parlor will be provided as their headquarters at 
the hotel. 

Note: The Pacific Coast Section officers and managers gave serious considera- 
tion to the question of holding the 1942 Spring Convention at Hollywood, and 
have decided to proceed with arrangements for the meeting. The motion picture 
industry plays an essential part from the exhibiting and engineering viewpoint in 
upholding the morale of the general and theater-going public in the present crisis, 
and accordingly the Convention and Local Arrangements Committees are por- 
ceeding with their plans. However, if later deemed advisable in the National 
interest, the Convention will be subject to cancellation thirty days prior to the 
announced Convention dates. 

W. C. KUNZMANN, 

Convention Vice-P resident 



BACK NUMBERS OF THE TRANSACTIONS AND JOURNALS 

Prior to January, 1930, the Transactions of the Society were published quar- 
terly. A limited number of these Transactions are still available and will be 
sold at the prices listed below. Those who wish to avail themselves of the op- 
portunity of acquiring these back numbers should do so quickly, as the supply 
will soon be exhausted, especially of the earlier numbers. It will be impossible 
to secure them later on as they will not be reprinted. 



1924 



1925 



No. 
19 

20 
21 
22 
23 
24 



Price 
$1.25 
1.25 
1.25 
1.25 
1.25 
1.25 



1926 



1927 



No. 

25 

26 

27 

28 

29 

32 



Price 
$1.25 
1.25 
1.25 
1.25 
1.25 
1.25 



1928 



1929 



No. 
33 

34 
35 
36 
37 
38 



Price 
$2.50 
2.50 
2.50 
2.50 
3.00 
3.00 



Beginning with the January 1930, issue, the JOURNAL of the Society has been 
issued monthly, in two volumes per year, of six issues each. Back numbers of 
all issues are available at the price of $1.00 each, a complete yearly issue totalling 
$12.00. Single copies of t*he current issue may be obtained for $1.00 each. 
Orders for back numbers of Transactions and JOURNALS should be placed through 
the General Office of the Society and should be accompanied by check or money- 
order. 



SOCIETY SUPPLIES 

The following are available from the General Office of the Society, at the prices 
noted. Orders should be accompanied by remittances. 

Aims and Accomplishments. An index of the Transactions from October, 
1916, to December, 1929, containing summaries of all articles, and author and 
classified indexes. One dollar each. 

Journal Index. An index of the JOURNAL from January, 1930, to December, 
1935, containing author and classified indexes. One dollar each. 

Motion Picture Standards. Reprints of the American Standards and Recom- 
mended Practices as published in the March. 1941, issue of the JOURNAL; 50 cents 
each. 

Membership Certificates. Engrossed, for framing, containing member's name, 
grade of membership, and date of admission. One dollar each. 

Journal Binders. Black fabrikoid binders, lettered in gold, holding a year's 
issue of the JOURNAL. Two dollars each. Member's name and the volume 
number lettered in gold upon the backbone at an additional charge of fifty cents 
each. 

Test- Films. See advertisement in this issue of the JOURNAL. 



JOURNAL OF THE SOCIETY OF 
MOTION PICTURE ENGINEERS 

VOLUME XXXVIII APRIL, 1942 

CONTENTS 

Color Television P. C. GOLDMARK, 

J. N. DYER, E. R. PIORE, AND J. M. HOLLYWOOD 311 

New Stop Bath and Fixing Bath Formulas and Meth- 
ods for Their Revival J. I. 
CRABTREE, L. E. MUEHLER, AND H. D. RUSSELL 353 

Book Review 373 

Officers and Governors of the Society 374 

Committees of the Society 377 

Constitution and By-Laws of the Society 382 

Fifty-First Semi- Annual Convention, Hollywood, 
Calif., May 4-8, 1942 393 

Society Announcements 397 

(The Society is not responsible for statements of authors.) 



JOURNAL OF THE SOCIETY OF 
MOTION PICTURE ENGINEERS 

SYLVAN HARRIS. EDITOR 
Board of Editors 

ARTHUR C. DOWNES, Chairman 

JOHN I. CRABTREE ALFRED N. GOLDSMITH EDWARD W. KELLOGG 

CLYDE R. KEITH ALAN M. GUNDELFINGER CARLETON R. SAWYER 

ARTHUR C. HARDY 

Officers of the Society 

*President: EMERY HUSE, 

6706 Santa Monica Blvd., Hollywood, Calif. 
* Past-President: E. ALLAN WILLIFORD, 

30 E. 42nd St., New York, N. Y. 
"Executive Vice-President: HERBERT GRIFFIN, 

90 Gold St., New York, N. Y. 
** Engineering Vice-President: DONALD E. HYNDMAN, 

350 Madison Ave., New York. N. Y. 
*Editorial Vice-President: ARTHUR C. DOWNES, 

Box 6087, Cleveland, Ohio. 
** Financial Vice-President: ARTHURS. DICKINSON, 

28 W. 44th St., New York, N. Y. 

* Convention Vice-President: WILLIAM C. KUNZMANN, 

Box 6087, Cleveland, Ohio. 
* Secretary: PAUL J. LARSEN, 

1401 Sheridan St., N. W., Washington, D. C. 
"Treasurer: GEORGE FRIEDL, JR., 

90 Gold St., New York, N. Y. 

Governors 

*MAX C. BATSEL, 501 N. LaSalle St., Indianapolis, Ind. 
**FRANK E. CARLSON, Nela Park, Cleveland, Ohio. 
*JOHN G. FRAYNE, 6601 Romaine St., Hollywood, Calif. 

* ALFRED N. GOLDSMITH, 580 Fifth Ave., New York, N. Y. 
**EDWARD M. HONAN, 6601 Romaine St., Hollywood, Calif. 

*I. JACOBSEN, 177 N. State St., Chicago, 111. 
**JOHN A. MAURER, 117 E. 24th St., New York, N. Y. 
*LOREN L. RYDER, 5451 Marathon St., Hollywood, Calif. 

* Term expires December 31, 1942. 
** Term expires December 31, 1943. 



Subscription to non-members, $8.90 per annum; to members, $5.00 per annum, included 
in their annual membership dues; single copies, $1.00. A discount on subscription or single 
copies of 15 per cent is allowed to accredited agencies. Order from the Society of Motion 
Picture Engineers, Inc., 20th and Northampton Sts., Easton, Pa., or Hotel Pennsylvania, New 
York, N. Y. 

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 

General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 

Entered as second-class matter January 15, 1930, at the Post Office at Easton, 

Pa., under the Act of March 3. 1879. Copyrighted, 1942, by the Society of Motion 

Picture Engineers, Inc. 



COLOR TELEVISION* 
P. C. GOLDMARK, J. N. DYER, E. R. PIORE, AND J. M. HOLLYWOOD 5 



Summary. A brief history of color television and the reasons leading up to the 
CBS Color Television System are presented, and a general theory for color television, 
including color, flicker, and electrical characteristics is also given. 

Equipment for color television transmission and reception has been designed and 
constructed based on these principles. 

INTRODUCTION 

Much of the significance of color in television is striking even to 
the casual observer. Aside from the most obvious effect, namely, 
that color introduces a sense of reality and a lifelike quality into the 
picture, comparison of a color television picture with the corre- 
sponding black-and-white image makes it apparent that not only 
are small objects more perceptible but outlines in general seem to be 
more clearly defined. As has been experienced with other media, 
color in television also seems to introduce a certain perception of 
depth. This is partly due to the increased ability of color to re- 
produce the contrasts and shadows as well as highlights and reflec- 
tions in different hues, while the degree of color saturation, which is 
a function of distance, especially outdoors, strongly enhances the 
three-dimensional quality. 

Effects such as those mentioned here became apparent immediately 
after initial experimentation with color television and were en- 
couraging enough to warrant an extensive investigation of that 
field with the objective of producing a practicable color television 
system. 

At the outset it was realized that the transmission and reception 
of live objects as well as motion picture film in natural color entailed 
the use of a trichromatic system. 

* Presented at the 1941 Fall Meeting at New York, N. Y., and at the New 
York Meeting of the Institute of Radio Engineers, October 3, 1941 ; received 
Octobers, 1941. 

** Columbia Broadcasting System, New York, N. Y. 

311 



312 GOLDMARK, DYER, PIORE, AND HOLLYWOOD [j. s. M. P. E. 

HISTORY 

Before proceeding further a brief summary of color television de- 
velopments in the past will be presented. In view of the large num- 
ber of proposals for color television systems, only those are mentioned 
which to the knowledge of the authors have been demonstrated. 

A complete color television bibliography, which is almost identical 
to that compiled by Panel 1 of the National Television System Com- 
mittee, is appended to this paper. 

Color television was demonstrated for the first time in July, 1928, 
by John L. Baird in England. Both at transmitter and receiver a 
three-spiral scanning disk was employed. Each of these spirals con- 
sisted of a succession of holes which were covered with red, green, 
or blue filters, scanning the picture completely in the three primary 
colors. At the transmitter photocells were employed, while at the 
receiver two gas-discharge tubes controlled by a commutator were 
used. One of the tubes was filled with neon and acted on the red 
spiral, while the other tube, filled with a mixture of helium and mer- 
cury vapor, appeared through the blue and green spirals. The trans- 
mission employed a band width of the order of 10 kc and the pictures 
corresponded to a number of lines somewhere between 20 to 30 per 
frame. 1 

In July, 1929, the Bell Telephone Laboratories in New York dem- 
onstrated a three-color television system employing three inde- 
pendent channels. The live pick-up equipment consisted of three 
banks of cells with the three primary color responses. A flying spot 
scanned the object and a scanning disk served on the receiving end 
to reconstitute the image. Three discharge tubes furnishing red, 
green, and blue light and superimposed by mirrors behind the scanning 
disk served as the light-source. 2 

It is interesting to note that while the Bell Laboratories employed 
a three-channel system which obviously occupies three times the 
frequency spectrum over the corresponding black-and-white picture 
and requires three times the facilities, Baird, though similarly re- 
quiring three times the frequency space, employed rotating filters 
and was thus the first to demonstrate the sequential, additive method 
of color in television. 

Early in 1938 John L. Baird in England demonstrated a 9 X 12-ft, 
120-line color television picture using sixfold interlacing, employing 
a flying spot, mirror drum, and rotating filters at the transmitter. 
At the receiving end also a mirror drum was employed, rotating at the 



April, 1942] COLOR TELEVISION 313 

rate of 6000 rpm and using a Kerr cell as modulator in conjunction 
with rotating color-filter slots. 

In July, 1939, a demonstration with similar transmitting equip- 
ment was reported. 3 At the receiver there was a projection cathode- 
ray tube combined with a rotating color-filter. The system was a 
two-color system using alternately orange and blue-green filters. 
The color picture frequency was 16 2 / 3 per second employing 102 
lines. 

Finally, on August 28, 1940, a three-color, high-definition tele- 
vision system employing electronic scanning both at the transmitter 
and at the receiver was broadcast for the first time over Station 
W2XAB in New York City. The subject of transmission was 
motion picture color-film. Soon after, live pick-up employing the 
same trichromatic system was demonstrated. 

This paper will deal largely with the color television system 
demonstrated on those two occasions, as well as its subsequent de- 
velopment up to the summer of 1941. Beginning June 1, 1941, 
daily color transmissions over Station WCBW of the Columbia 
Broadcasting System inaugurated a field- test period for the purpose 
of determining the practicability of color television as presented in 
this paper. 

CBS COLOR TELEVISION 

At the outset of research activities in color television a number of 
conditions were set down upon fulfillment of which depended the 
success of practicable color television. These were: 

(1) For a given band-width the loss in monochromatic definition due to the 
introduction of color should not be excessive. 

(2} The system should be based upon three primary colors. 

(5) Within any given band- width the performance of the color system decided 
upon should give at least as much satisfaction as the corresponding black-and- 
white system. 

In the color television system under discussion the following terms 
will be used : 

Color- Field Frequency. The highest vertical scanning frequency employed in 
the system. 

Color- Frame Frequency. Color repetition time per second, i. e., trichromatic 
repetition rate per second, corresponding to the color field frequency divided by 
three. 

Color-Picture Frequency. Number per second of the coincidence of one and the 
same primary color with one and the same area of the image. 



314 



GOLDMARK, DYER, PIORE, AND HOLLYWOOD [j. S. M. P. E. 



Frame Frequency, Identical to term used in monochromatic television, i. e., 
completion of the scanning of the entire picture area per second in black and 
white. 

Before the choice for a final system was narrowed down, several 
alternatives were considered. These all had in common sequential, 
additive color scanning where the primary color impulses of varying 
ratio, following in rapid succession, are integrated by the observer's 
eyes. The three primary colors employed were red, blue, and green, 
the characteristics of which will be discussed later. Rotating color 
disks or drums in front of the pick-up device and the receiving tube, 
suitably synchronized and phased, produced the color analysis at the 
transmitter and the synthesis at the receiver. 

The various systems combining different interlace ratios, color- 
fields, color-frame, and color-picture frequencies as well as lines are 
tabulated in Table I. 



System 



TABLE I 

Characteristics of Various Systems 
123 



Color-fields (c} 
Color-frames 
Frames (/) 
Color-pictures 
Interlace ratio (c/f) 
Lines per frame, corresponding 
to 441 black-and-white, near- 
est practical value 
Horizontal (line) frequency 
Lines per frame, corresponding 
to 525 black-and-white, near- 
est practical value 
Horizontal (line) frequency 
Color break-up conditions 
Interline flicker conditions 
Picture flicker conditions 



60 


120 


120 


180 


120 


20 


40 


40 


60 


40 


30 


120 


60 


45 


30 


10 


40 


20 


15 


10 


2 : 1 


1 : 1 


2 : 1 


4 : 1 


4 : 1 


441 


240 


315 


350 


441 



13,230 28,800 18,900 15,750 13,230 
525 260 375 450 525 



15,750 51,200 22,500 20,250 15,750 

Unsatis Satis Satis Satis Satis 

Unsatis Satis Satis Doubtful Unsatis 

Unsatis Satis Satis Satis Satis 



System 1, as can be seen, is a straight adaptation of the black-and- 
white standards. Though the monochromatic definition was un- 
impaired, the resulting flicker due to the low color-frame frequency 
(20) was intolerable even at low illumination values. The effect, 
known as "color break-up," which is purely a physiological one and 
increases with decreasing color-frame frequency, was objectionable. 
A white object which moved across the screen with sufficient velocity 



April, 1942] 



COLOR TELEVISION 



315 



would show red, blue, and green color fringes. Only empirically 
could it be determined at which color-frame frequency the color 
break-up for the most commonly transmitted objects in motion be- 
came negligible. 

In System 1 the color-picture frequency is 10, which gives rise to 
interline flicker. It will be obvious when considering the subsequent 
systems that for a constant frame-frequency the color-picture fre- 
quency also remains the same. This is an important factor when 
considering interlace ratios higher than 2. 

System 2 employs sequential scanning in order to eliminate inter- 
line flicker. The same frequency is 120, and thus the color-field 
frequency becomes 40 per second. Flicker and color break-up con- 
ditions are satisfactory. However, the loss in definition is excessive. 

RED BLUE GREEN RED BLUE GREEN 



roi fiR nn n IMTTRY/AI (SEC.) 


i 










120 




60 




40 



20 

FIG. 1. Diagrammatic representation of CBS System 3. 

It became evident that in order to increase definition interlacing 
had to be introduced. This led to a system having a color-field 
frequency of 120, a color-frame frequency of 60, and color-picture 
frequency of 20 per second. Due to a 2 : 1 interlace ratio the frame- 
frequency remains at 60 per second and the number of lines at 343. 
This system gave freedom from flicker with screen brilliancies up to 
2 apparent foot-candles and showed no interline flicker. It was 
subsequently chosen as the most satisfactory compromise for the 
present 6-Mc band at the same time increasing the number of lines 
to 375 which corresponds to the 525 lines used in monochromatic 
television. 

System 4 is a compromise between Systems 3 and 5 inasmuch as 
the color-picture frequency is 15 with a corresponding color-frame 
frequency of 60 and color-field frequency of 180 per second. The 
frame-frequency of 45 per second will permit a number of lines ap- 
proximating 525-S/30/45. In this system flicker even at the highest 



316 



GOLDMARK, DYER, PIORE, AND HOLLYWOOD [j. S. M. P. E. 



brilliancies is eliminated; however, the interline flicker still appears 
somewhat excessive. 

System 5 uses the same horizontal scanning frequency as mono- 
chromatic television; however, it utilizes quadruple interlacing to 
increase the field-frequency to 120 per second. Thus flicker condi- 
tions are satisfactory and resolution is excellent. However, due to 
the low color-picture frequency (10 per second) interline flicker ap- 
pears excessive. 




2 3 



FIG. 2. Unified trichromatic coefficient diagram (color triangle). 

In order to avoid the so-called "line crawling" effect the quadruple 
interlacing in Systems 4 and 5 is of the staggered type where the 
sequence of lines is 1, 3, 2, 4 instead of 1, 2, 3, 4. 

Conditions in these two systems are aggravated by the fact that 
the color-field frequency of, respectively, 180 and 120 per second, 
being a multiple of the power-supply frequency, would show a dis- 
tinct breakdown of the line-structure emphasizing a raster of ap- 
proximately 100 lines in case 60-cycle components were not completely 
eliminated from the vertical scanning. 



April, 1942] 



COLOR TELEVISION 



317 



The final decision in favor of System 3, with System 4 as a close 
second, had to be made in view of the discouraging results, con- 
firmed by other experimenters, in attempting to reach a satisfactory 
solution of the quadruple interlacing problem in general. 

In Systems 1,3,4, and 5 the number of color-fields over frames per 
second was an even number. If in a trichromatic system all areas of 




FIG. 3. Color characteristics at the receiver: filters 
Nos. 47, 58, and 26 combined with the phosphor. The 
fourth curve represents an average phosphor. 



the image are to be scanned in all three primary colors, then the 
following conditions must be fulfilled : 



= 3n 



where c = color-fields per second 
/ = frames per second 
n = any whole number, 0, 1, 2, 3, etc. 



For sequential scanning n and thus c/f = 1, for n = 1, c/f 
becomes either 2 or 4, which corresponds, respectively, to double 
and quadruple interlacing. Fig. 1 is a diagrammatic representation 
of System 3. 



318 



GOLDMARK, DYER, PIORE, AND HOLLYWOOD [j. S. M. p. E. 




4000 



FIG. 4. Color characteristics at the transmitter: 
these curves are solutions of equation 9 on the basis of 
receiver characteristics represented in Fig. 3. 




FIG. 5. Color characteristics at the transmitter: 
the curves combine the standard dissector, the carbon 
arc, 2 mm of Corning glass No. 978, and Wratten niters 
Nos. 25, 47, and 58. 



April, 1942] 



COLOR TELEVISION 
COLOR 



319 



The color television method under discussion is based on the eye's 
retentivity of light of all colors and its ability to recognize mixtures 
of several colors as a single one. Because of the fact that theoreti- 
cally all colors are reproducible by a suitable set of three primaries, 
a trichromatic color system was chosen as the basis for color tele- 
vision. 




FIG. 6. Color characteristics at the transmitter: 
the curves combine the daylight dissector attenuated 
with the carbon arc, 1 mm of Corning glass No. 978, and 
Wratten filters Nos. 25, 47, and 58. 

The experimental fact that the stimulant E(\) produces a sensa- 
tion equivalent to a mixture of three primaries can be expressed as : 

w 2 (X)7. 



where the w's are coefficients and the I's are the three primary colors. 
Before proceeding to the discussion of the application of equation 
1 and the choice of primaries in reproducing color in television, it is 
best to introduce some concepts used in standard colorimetry. It 
can be shown that although the w's can be determined experimen- 
tally, there is nothing unique about a given set of primaries and their 
respective coefficients. The present practice is to use coefficients 
determined by the International Commission on Illumination in 
1931, for the standard observer. 4 



320 



GOLDMARK, DYER, PIORE, AND HOLLYWOOD [j. s. M. p. E. 



These coefficients, known as the ICI tristimulus values of the 
spectrum colors, are usually designated by x(\) dominant in the red 
region of the spectrum, y(\) in the green, and z(X) in the blue region. 
The tabulated values and the curves for these tristimulus values 
can be found in several places. 5 The primaries associated with 
x, y, and z are fictitious in the sense that they have no physical 
counterpart. However, they never assume negative values. Pur- 
posely y(\) was chosen to coincide with the luminosity curve or the 
standard visibility function and thus represent the energy of the 
light-source in terms of the eye or the brilliance of the color. 

With the aid of the tristimulus values x, y, z, a graphic representa- 
tion of all colors can be constructed. This is known as the color 





FIG. 7. Color characteristics at the transmitter: 
Wratten filters Nos. 27, 47, and 57 combined with day- 
light fluorescent light-source as used with orthicons. 

triangle, or the unified trichromatic coefficient diagram, and is 
shown in Fig. 2. The locus of pure spectral colors is shown with a 
solid curve. This curve is obtained by the so-called unified coef- 
ficients, i. e.: 



x = 



x + y 



y = x + y + z, z = x + y 



; as a result x -f- y + z = 1 



To determine the tristimulus values for a light-source E(\), an 
integration must be performed, i. e.: 

(2) 



, Y' = f 

For these values the unified trichromatic coefficients are: 
X' Y' Z' 



X = 



X' + Y' + Z" 



X' + F' + Z" 



X' + Y' + Z' 



April, 1942] 



COLOR TELEVISION 



321 



The light-source (X) can now be represented by a point in the color 
triangle. The operation shown in equations 2 can be carried out 
on luminants of any spectral distribution, such as fluorescent ma- 
terials, illumination sources with or without color-filters, etc., after 
which their respective positions in the color triangle can be deter- 
mined. The second equation in the group determines the luminosity 
of the color. In Fig. 2, point w represents white and greys, i. e.: 

X = Y = Z 
The gamut of colors which a given set of primaries will reproduce 



A CIRVE IN THE 



OAYLIGH 

FLUORESCE 



\ 



FIG. 8. Comparison of daylight fluorescent light-source 
with actual daylight. 



can be determined with the aid of the color-diagram. (The pri- 
maries in this discussion are used only with positive coefficients.) 
Only those colors enclosed within the triangle, the corners of which 
are represented by the primaries, can be reproduced by the afore- 
mentioned primary colors. Two such triangles are drawn in Fig. 2. 
In order to reproduce the largest possible gamut, the three primary 
colors must be so chosen that the resulting triangle encloses most of 
the colors commonly used. Usually this entails a compromise, with 
a sacrifice of some of the blue-green region. 

The limitations encountered in any two-color system are obvious 



322 GOLDMARK, DYER, PIORE, AND HOLLYWOOD [j. s. M. P. E. 

when examining Fig. 2. Employing only two primaries, the colors 
that can be reproduced are only those found along the straight line 
joining the location of the two primaries in the color triangle. 

Color Characteristics at the Receiver. In the television system 
under discussion the primaries at the receiver are determined by the 
color-filters, red, green, and blue, and the fluorescent material in the 
tube. The largest gamut of colors is produced with primaries which 
fall on the locus of monochromatic colors in the color triangle. One 
such choice is red 7000 A, green 5350 A, and blue 4000 A, shown in 
Fig. 2. 

Unfortunately, monochromatic primaries can be obtained only 
at the sacrifice of light-intensity. Thus one finds that in television, 
as in certain color-reproducing processes, a compromise must be 
found between light-intensity and the best choice of primaries. In 
addition, there is a restricted choice in available phosphors. The 
decay time of the fluorescent powder used in the receiving tube must 
be such that its intensity becomes negligible after one color-field 
period. 

Of the commercially available phosphors the zinc and cadmium 
sulfides possess sufficient luminescent efficiency and also satisfy the 
decay requirements. The luminescent spectrum of the phosphor 
must cover the entire range of the three filters in order to provide a 
light-source for each primary. The precise character of the spectrum 
desired is contingent upon the choice of filters. The most desirable 
characteristic would be to furnish maximum light in those blue, 
green, and red regions that fit the maximum transmission portions 
of the blue, green, and red filters. Commercial tubes usable for 
color television employ, for the most part, two component mixtures 
utilizing a zinc-sulfide with a spectral emission maximum in the blue 
and blue-green region, and a zinc-cadmium sulfide with a maximum 
in the yellow and yellow-red region. 7 The spectral curve of such a 
mixture is shown in Fig. 3. 

An inexpensive source of spectroscopically reproducible filters, 
with a wide color-selection, is the Wratten series, available in gelatin 
or acetate stock. The choice of the filters is determined by the wave- 
length at which the maximum transmission occurs, the width of the 
transmission band, and the total transmission. It is desirable to 
have filters and phosphor so chosen as to produce white corresponding 
to 5500 K with equal signals on the grid of the picture tube during 
the red, blue, and green periods. Thus if a white surface of 5500K 



April, 1942] 



COLOR TELEVISION 



323 



FIG. 9. Solution of equation 9, including a decay of 1 / 6 
from one color-field to the next in the orthicon. 



* 20 



GREEN 



FIG. 10. Solution of equation 9, including a decay of 
Vio from one color-field to the next in the orthicon. 



324 GOLDMARK, DYER, PIORE, AND HOLLYWOOD [j. s. M. P. E. 

is transmitted, it should be received as the same shade of white and 
also should be identical to the receiver's own color when operated 
without a signal. 

The filters finally chosen for use at the receiver were Wratten No. 
26 for red, No. 47 for blue, and No. 58 for green. The emission 
curves for the phosphor mixture used for the experimental tubes com- 
bined with filters Nos. 47, 58, and 26 are given in Fig. 3. The resul- 
tant blue, green, and red primaries yield a new color-triangle repre- 
sented with broken lines in Fig. 2 ; the location of the new primaries 
is marked with the corresponding filter numbers. 

A satisfactory method of specifying the color composition of the 
phosphor, which is a mixture of blue, yellow, and orange zinc and 
zinc-cadmium sulfides, without resorting to actual spectral curve 
data was to specify the transmission through filters Nos. 47, 58, and 
26 as recorded with a Weston photocell No. 2. The relative values 
which proved satisfactory were No. 47, 1.0; No. 58, 1.25; and No. 
26, 0.75. Of the commercially available tubes, the General Electric 
and Baird have been found quite useful. RCA Radiotron and Na- 
tional Union have supplied special tubes that have proved satis- 
factory. In the CBS laboratories, tubes have been built using a 
phosphor mixture containing 60 per cent of a blue zinc-sulfide, 35 
per cent of a yellow-green zinc-cadmium sulfide, and 5 per cent of a 
reddish zinc-cadmium sulfide. The final color characteristic of the 
tube, however, depends to a large extent upon the various processing 
schedules. 

The range of colors obtainable with this choice of phosphors and 
filters (shown in Fig. 2 with a dotted triangle) is as large as is en- 
countered in color photography. The white produced with three 
equal signals appears somewhat bluish. Very recently, however, 
satisfactory "white" tubes were made in the laboratories, which 
show consistently good color-characteristics and permit transmitter 
operation with equal blanking pulses. Spectral curves for these 
tubes are in preparation and will be available shortly for standardiza- 
tion purposes. 

Transmitter Color-Characteristics. While the performance of the 
receiver was based upon the theory of color vision, the study of the 
color-characteristics at the transmitting end of the system had to 
oe guided by the desirability of producing all colors encountered in 
nature. At the receiver three properly chosen narrow bands in the 
spectrum were sufficient; at the transmitter, the bands must be wide 



April, 1942] COLOR TELEVISION 325 

enough and sufficiently overlapping to produce a signal from every 
color. The exact character of the three spectral curves at the trans- 
mitter is determined by the filter and phosphor combination at the 
receiver. 

The spectral characteristics at the transmitter are composed of the 
spectral sensitivity of the pick-up tube in the camera, the transmis- 
sion curves of the filters, and the spectral emission of the light- 
source. The filters and light-source must be so chosen as to produce 
negligible signals in the infrared (beyond 7000 A) and the near ultra- 
violet (below 4000 A) regions of the spectrum. 

The general relationship between the color-characteristics of the 
transmitter and the receiver can be derived from an analysis de- 
veloped by Hardy and Wurzburg in connection with photographic 
reproduction in color. 6 

The tristimulus values for any object at the transmitter, the spec- 
tral reflection of which is represented by (X), was given by equation 
2. At the receiver the tristimulus values are represented by : 

x " = r 



bf Q P(\)F b (\}z(\)d\ 
(5) 

The coefficients r, g, and b are proportional to the light-intensities 
at the picture tube associated with signals generated at the camera 
through the red, green, and blue filters. P(X) is the spectral emis- 
sion of the phosphor. F r (\), F g (\), and F b (\) are the spectral 
transmissions of the red, green, and blue filters at the receiver, and 
x(\), y(\), and z(X) are the ICI tristimulus values. 

The r, g, and b are some functions of the light incident at the 
photosensitive member of the camera tube, i. e.: 

r = f(S r ), g = f(S ), b = f(S b ) (4} 

Simplified for further discussion, equation 3 can be written as: 

X" = ra n + ga n + ba u 

Y" = ran + ga 2Z + ba zs (5) 

Z" = ra 3 i 4- ^32 + ba S3 

For the system to reproduce color faithfully it is necessary that 
equations 2 and 3 be related linearly, i. e.: 

X' = kX"; Y' = kY"; Z' = kY k (6} 

where k is a constant. 



326 



GOLDMARK, DYER, PIORE, AND HOLLYWOOD [j. s. M. P. E. 



The S r) S g , S b of equation 4 is that portion of the light at the 
transmitter which is useful in producing the electrical signal. It 
can be resolved into three components : A (X), the spectral sensitivity 
of the photosensitive member of the camera tube, the spectral at- 
tenuation of the red, green, and blue filters at the camera tube, 
T r (\), T g (\), and T b (\) and the spectral distribution of the object 
E(\) being televised. E(X) will depend upon the light-source used 
to illuminate the object. The S's can thus be written as: 



s r ^ 

S = 
S b = 



d\ 



(7) 



3 4 56789 10 

FOOT CANDLES 



FIG. 1 1 . Relationship between critical flicker frequency and picture brilliance. 

The actual functional relation between r and S r , etc., depends 
upon the relationship between the light input and the signal output 
at the camera, on the relationship between the voltage on the grid 
of the receiving picture, and the light output from the phosphor; 
it further depends upon the electrical characteristics of the terminal 
equipment, the transmitter proper, and the receiver. 

Equations 2 and 3 permit the determination of 

T r (\)A(\), T (\)A(\), and T b (\)A(\) 

for a given light-source. Once P(\)F r (\), P(\)F g (\), and P(\)F r (\) 
have been chosen and provided it is assumed that r = S r , g = S g , 



April, 1942] 



COLOR TELEVISION 



327 




FIG. 12. Dissector block diagram. 



sa iss 




FIG. 13. Orthicon block diagram. 



328 GOLDMARK, DYER, PIORE, AND'HOLLYWOOD [j. s. M. p. E. 




FIG. 14. Orthicon camera on tripod; direct pick-up color-camera. 




FIG. 15. Orthicon color-camera; inside view, showing filter-drum 
with synchronous driving motor. 



April, 1942] COLOR TELEVISION 329 

and b = S b . This in effect signifies that the gamma of the overall 
system is unity. If equation 5 is substituted in 2 the new equation 
is in turn substituted with equation 7 into 3, and the integral sign 
and (X) are cancelled heuristically from both sides, the resulting 
relationship is : 



;y(X) = 

5(X) = T r (\)A(\)a 3l 

The color-response at the transmitter is contained in the quantities 
T r (\)A(\), T g (\)A(\), and T b (\)A(\). In equation 8 the coef- 
ficients a mn are known constants that are evaluated in terms of color- 
response at the receiver and the ICI tristimulus values. Equation 
8 can thus be written as: 



= (a 22 a 33 a 23 a 32 )jc(X) + (a i3 a 32 
cTg(\)A(\) = (023^31 a 2 ia ss )x(\) + (a n a Z 3 
c7fc(X)4(X) = (a 2 ia 32 a 2 23i)*(X) + (aiya n a n a 32 );y(X) 

(9} 

where c is a constant. The solution of 9 in terms of receiver color- 
response as represented in Fig. 3 is shown in Fig. 4. The curves in 
these figures display the usual characteristics found in all color- 
reproduction problems the existence of negative color values for 
perfect matching. The present color television system has no 
mechanism to introduce these negative values; however, they are 
being partially compensated for with the aid of the color mixer. 

The above analysis is valid for the color-transmission of live scenes 
as well as for color-slides or motion pictures, provided the latter two 
contain no color distortion. Once the photographic or printing proc- 
esses have introduced certain distortions, the filters at the trans- 
mitter must be so chosen as to compensate, if possible, for these 
specific anomalies. 

Camera Tubes. Before comparing the results contained in Fig. 4 
with actual operating conditions it is best to consider briefly the 
color-characteristics of the tube used at the transmitting end. They 
are of two types : the dissector, used for slides and motion pictures, 
and the orthicon, used in the studio and in outdoor pick-ups. One 
of the problems in dissector operation is the elimination of signals 
produced with infrared radiation. The infrared contaminates all 
colors as it passes freely through the red, blue, and green filters. 
Originally the standard dissector was used with a cesiated silver oxide 
cathode surface. This surface has a minimum color-response in the 



330 



GOLDMARK, DYER, PIORE, AND HOLLYWOOD [j. S. M. p. E- 



green portion of the spectrum while it shows rising tendencies toward 
both the blue and the infrared regions. In order to utilize this tube 
a carbon arc was used as a light-source, combined with the infrared - 
absorbing Corning glass filter No. 978, 2 mm thick. The filters 
that were used in this set-up were Wratten Nos. 47, 58, and 25. Fig. 
5 is a graphic representation of the results. The dissector used at 
present is the so-called daylight dissector (developed especially for 
color television by the Fransworth Television and Radio Corporation) 
with a maximum in the green portion of the spectrum falling off 
toward the blue and the red end. This dissector was used also with 




FIG. 16. Color-mixer amplifier. 

a carbon arc but with a Corning filter No. 978 only 1 mm thick. 
The color-filters were again Nos. 47, 58, and 25. In both cases a 
water-cell was used to protect the slides and film. Fig. 6 gives the 
operating conditions for the daylight dissector. There is no question 
of the superiority of this type of tube over the standard dissector 
for this work. The signal-to-noise ratio is improved partly because 
of the greater photoelectric response in the pertinent portion of the 
spectrum and also because of the reduction in thickness of the 
infrared filter. 

The spectral characteristics of the orthicon have not been mea- 
sured under operating conditions. However, from measurements of 
the signal obtained through various Wratten filters it can be stated 



April, 1942] 



COLOR TELEVISION 



331 



very generally that the spectral response of the color orthicon does 
not conform to the standard cesiated silver oxide surface. It was 
found that specifications of the color characteristics of the orthicons 
used can be summed up in terms of the signal obtained through 
Wratten niters Nos. 47, 57, and 25. The ratio of the signals through 
Nos. 47 to 57 using a daylight fluorescent light-source is approxi- 
mately 1.25, and the ratio of No. 47 to No. 25 is 0.85. To obtain a 
general view of the color characteristics at the transmitter a daylight 
fluorescent light-source attenuated through filters Nos. 47, 57, and 
25 has been plotted in Fig. 7. The inclusion of the characteristics of 
the orthicon in this figure will reduce the red curve roughly by 30 



TIME 


120 


120 


120' 


120 


120 


756 


RED 


~~~ 






"~~ 






KEYING PULSE 














BLUE 











| 




KEYING PULSE 














GREEN 
KEYING PULSE 

















COLOR TRANSMITTED: RED. BLUE, GREEN, RED, BLUE, GREEN, ETC. 

FIG. 17. Color-mixer pulse diagram. 

per cent. It may shift the maxima slightly but will not change the 
limits of the individual curves. Thus the extent of filter overlapping 
remains the same. 

Lighting. The color-characteristics at the transmitting end of the 
system depend to a large extent upon the illumination source. For 
slides and motion pictures the carbon arc is used and the filters are 
chosen to match the carbon arc. For direct pick-up it is desirable 
to have a light-source such as not to necessitate the change of filters 
when the camera is moved outdoors. A good approximation to 
such a light-source can be obtained with incandescent lamps where 
the infrared is attenuated with properly chosen glass filters, such as 
Aklo glass. However, this type of source is very inefficient. Fluores- 
cent lamps of the daylight type are a fair approximation to the 
requirements. The lamps contain 30-watt bulbs mounted in spe~ 



332 GOLDMARK, DYER, PIORE, AND HOLLYWOOD [j. s. M. P. E. 

cially designed reflectors, developed for color television to give the 
maximum light-flux for a minimum ceiling area. A single unit is 
shown in Fig. 31 and a bank in Fig. 32. 

Fig. 8 gives the spectral emission curve for the daylight fluorescent 
lamps. The spectral distribution of light as received from the sun on 
a horizontal plane under various cloud conditions and at different 
times of day also is shown. 8 

It can be seen that the daylight fluorescent lamps are down in the 
red and slightly down in the blue portions of the spectrum as com- 
pared to outdoor illumination. It was found that when switching 
from studio pick-up to outdoor scenes it is necessary to reduce the 
red signal by means of the electrical color mixer. 

Comparison of Figures. The calculations culminating in Fig. 4 
have been purposely kept in general terms using a light-source of 
uniform energy-distribution over the visible spectrum. Figs. 5 
through 7 take into consideration the actual light-source in order to 
show the signal magnitude through the filters and the attenuation 
of the infrared component. 

It follows that the comparison of Fig. 4 with Figs. 5 to 7 must be 
very general. The spectral characteristics of operational equipment 
lack the negative value. The maxima are roughly in the same region, 
and the limits of the blue and green curves, which are principally 
determined by the Wratten filters, correspond quite closely. A 
certain discrepancy exists in the reds, where in Fig. 4 the cut-off is 
at 5500 A and in Fig. 7 at 5800 A. When examining Fig. 5 carefully, 
one can see that monochromatic sources between 4000-4800 A will 
appear as the same blue, those between 5400-5800 A will appear as 
the same green, and those between 6300-6700 A will appear as the 
same red at the receiver. However, these imperfections are not too 
serious, since only rarely do objects reflect colors of such narrow 
bands. 

As has been pointed out previously, this color television system 
does not permit introducing the negative values (which are common 
to most color-reproducing systems) as shown in Fig. 4. Nevertheless, 
the color-mixer permits partial compensation by changing the ratio 
between signals r, g, and b. The system itself in unaware of whether 
the change takes place in the spectral characteristics of the light- 
source at the transmitting end or in light-emission of the receiver. 
Changing the relative amounts of the red, blue, and green signals 
electrically in effect changes the solution of equation 9. Thus the 



April, 1942] 



COLOR TELEVISION 



333 



manipulation of the color-mixer changes the ratio of positive to 
negative signal. Color deterioration resulting from the lack of 
negative signal is thus partially reduced. 

In this analysis, the persistence of a signal from one color-field to 
the following ones, as experienced in the orthicon under certain con- 
tions, has been neglected. Curves similar to those in Fig. 4 have 
been calculated, taking into account hangover of 1 : 5 and 1 : 10 
of the original signal into the next field for a color-sequence red- 
blue-green, and are shown in Figs. 9 and 10. The same receiver 




FIG. 18. Color-mixer pulse generator. 

color-characteristics as in Fig. 4 were assumed. The most out- 
standing feature of these curves is the increased amplitude and 
width of the negative portions. A hangover of 1 : 5 (Fig. 9) would 
require much broader filters at the transmitter and the color con- 
tamination would be appreciable. A hangover of 1 : 10 (Fig. 10) 
shows less deviation, as compared with conditions in Fig. 4, where no 
hangover was assumed. The blue filter is wider than in Fig. 4 and 
has a negative component in the red region. On the whole, however, 
Fig. 10 matches Fig. 4 reasonably well. 

The color-mixer has been used to compensate for different lighting 
conditions to correct for color contamination resulting from lack of 
negative signal, and thus also indirectly for hangover. There is no 



334 GOLDMARK, DYER, PIORE, AND HOLLYWOOD [j. S. M. p. E. 

assurance that all these corrections require the same adjustment. 
The mixer permits a partial removal of color contamination from a 
number of sources and in actual operations a compromise is made for 
the best overall effect. 

The analysis has been based on a linear relationship between light 
output at the receiver, i. e., a system with unity gamma, which con- 
ditions may not be obtained iwithj>resent equipment. 




FIG. 19. Filter disk for orthicon. 
FLICKER IN COLOR TELEVISION 

The well known Ferry-Porter 9 law states that the critical frequency 
is proportional to the logarithm of the illumination intensity. Porter 
was the first to establish the fact that the critical fusion frequency 
is independent of wavelength, provided the apparent brilliance re- 
mains constant. Based on this law he derived the well known color- 
intensity curve of the eye using a flicker photometer. It follows 



April, 1942] COLOR TELEVISION 335 

that with constant illumination over the visual spectrum the eye's 
sensitivity to flicker follows the color-intensity curve. 

In the sequential color television system under consideration the 
worst flicker condition would occur if only the primary color with the 
highest luminosity were received while the other two primary colors 
were suppressed completely. Such a condition hardly ever occurs in 
practice unless the color camera picks up a green object, the limits 
of its chromaticity being between 5400 A and 5800 A (see Fig. 5). 
The flicker frequency for this case would be 40 per second. An 
attempt will be made to calculate the maximum permissible bril- 
liance of such a green picture before flicker becomes perceptible. 

The validity of Talbot's law has been checked for all colors by 
Porter 10 and others. Thus, applying Talbot's law to the special case 
of color television as discussed here, the apparent brilliance of the 
image at the receiver would be : 

IA = Y (!(*)# (io) 

1 *Jo 

where T c is the duration of a complete color-cycle (color-frame) and 
T the duration of a color-field. L(t) is the decay function of the screen 
material. This is assumed to be exponential, with a luminosity not 
greater than one-tenth of the initial brilliance after the duration of 
one field period T. 

Thus for the transmission of green between 5400 A and 5800 A the 
apparent brilliance of the received picture becomes : 



I a = ^ C T L(t)dt (11) 

lc jo 



where the luminosity at the receiver L(t) is multiplied by the Y 
component of the unified trichromatic coefficients, representing the 
combination of the receiver's green filter and the screen material, as 
shown in the color diagram Fig. 2. 

For a picture tube made in the laboratories especially for color 
reception the following ratios of the luminosity values of the three 
color primaries were found: 

^^ = 23 and p* = 10.5 (12) 

* bluG * blue 

Substituting in formula 11 for T c = 1 / 40 and for T = V^o second 
and solving the integral, we obtain for the apparent brilliance which 
is furnished during the green period only: 

L(t)dt 



336 



GOLDMARK, DYER, PIORE, AND HOLLYWOOD [J. S. M. P. E. 



In order to determine at what apparent brilliance a 40-cycle 
television picture will just begin to show flicker we consult the curve 
shown in Fig. 11, which is from the article entitled "Television Image 
Characteristics" by Engstrom. 10 The curves were obtained by 
using a special film which corresponded to certain decay characteris- 
tics of screen materials. It was decided to choose film No. 1, the 
decay characteristic of which is sufficiently fast to correspond to a 
screen material usable in color television where at the end of one 
frame the screen brilliance decays practically to zero. Thus, ac- 
cording to Fig. 11, a repetition frequency of 40 per second will permit 
a screen illumination of 1.8 apparent foot-candles. 




FIG. 20. Orthicon color-camera; filter-drum assembly. 

So far we have considered the most unfavorable case from a flicker 
point of view. More favorable conditions will occur if white with 
three equal electrical impulses during the red, green, and blue periods 
is transmitted and received. 

In order to obtain the total apparent illumination at the receiver, 
formula 11 will be expanded into: 



/ = TTv 



L(t)dt 



(14) 



It has been assumed that the spectral characteristic of the phos- 
phor does not change during the decay. 

Substituting for Y g = 23Y b and Y r = 10.5 Y b from equation 14 we 
obtain : 



April, 1942] 



COLOR TELEVISION 
34.5 



' = 



337 

(IS) 



previously (13) it was found that 

_ 23 
L ~40 

and thus 

A, = 34.5 

In ~ 23 



rr 

?b L()dt 
Jo 




FIG. 21. Nine-inch color television receiver; 
front view. 

The meaning of this expression is that the apparent brilliance of 
the receiver screen increases only in the ratio of 34.5/23 when white 
is transmitted, even though it is produced by three equal electrical 
impulses, one during each color-field. Since in this case a light- 
impulse is received during each of the color-fields, flicker conditions 
improve rapidly. However, the apparent brilliance should not be 
higher than 1.8 times 34.5/23 which is 2.7 apparent foot-candles, 
if one wishes to make sure that in the singular case of the transmis- 
sion of a narrow band of green no flicker is present. This top value 
of 2.7 foot-candles is not a serious limitation. Present black-and- 



338 GOLDMARK, DYER, PIORE, AND HOLLYWOOD [j. S. M. p. E. 

white pictures with such a highlight brilliance give satisfactory 
viewing in darkness. The same black-and-white receiver, however, 
will not permit satisfactory viewing with surrounding illumination 
of any appreciable magnitude. 

Color television pictures produced with the aid of rotating niters 
do not deteriorate appreciably in surrounding illumination due to 
the fact that the room light which passes through the niters twice is 
attenuated by the square of the filter-loss factor, while the picture 
itself is attenuated only by the first power of the filter-factors.* As 




FIG. 22. Nine-inch color television receiver; top view, open. 

a result the unmodulated screen of a color receiver appears nearly 
black even in a well illuminated room. Thus the 2.7-foot-candle 
maximum brilliance will furnish a more satisfactory image than a 
conventional black-and-white receiver even with 1 0-f oot-candle high- 
light illumination. 

EQUIPMENT FOR COLOR TELEVISION 

Studio. Certain electrical requirements must be met by pick-up 
tubes if they are to be used in color television. It is important that 

* A black-and-white receiver with an equivalent neutral filter will resist room 
illumination to the same extent. 



April, 1942] COLOR TELEVISION 339 

the signals produced during any one color-field should not be adul- 
terated by a signal left over from a previous field. Storage-type 
camera tubes must, therefore, be designed so that the entire elec- 
trical charge on the mosaic is removed within one field period. 

A constant black level must be established in the camera tube, and 
spurious signals such as "shading" should be absent. The dissector 
is the only commercial camera tube that meets all the above require- 
ments, though its usefulness is limited to the transmission of film or 
slides due to its low sensitivity. 

The orthicon as modified for color television with lower mosaic 
capacity was developed through the cooperation of the RCA Radio- 




FIG. 23. Seven-inch color television receiver; front view. 

tron Division and has been found to produce very acceptable color- 
pictures with incident light of 150 foot-candles on the subject. A 
certain amount of "hangover," which may be defined as the amount of 
signal remaining on the mosaic after the scanning beam has com- 
pleted one field, appears to be unavoidable, but is troublesome only 
at lower light levels. The hangover occurs when the potentials on 
the mosaic are small enough to be within range of the random velocity 
distribution of the scanning beam. Under this condition the charge 
is not completely removed until the mosaic has been scanned several 
times. A lower mosaic capacity permits the voltage on the mosaic 
(for a given illumination) to build up to higher levels. 

It might be expected that difficulty would be experienced with 



340 



GOLDMARK, DYER, PIORE, AND HOLLYWOOD [j. s. M. P. E. 



interlaced scanning on a storage-type camera tube such as the orthicon. 
In an ideal storage tube, where the scanning spot is only one line 
wide, it will be apparent that at the end of one field-scanning period 
only one-half the lines will be scanned and the hangover will be 100 
per cent. Actually, in practice, hangover ratios of from 1 : 5 to 1 : 1 
are obtained with the orthicon, indicating that either leakage, 
defocusing, or other effects are present to such an extent as to remove 
most of the unscanned picture at the end of one field scansion. 




FIG. 24. Seven-inch color television receiver; top view, 
open. 

The gamma of a camera tube need not necessarily be unity, as 
correction may be made for any particular characteristic later on, if 
desired. In general, a television system employing a linear pick-up 
tube, such as an orthicon, will have an overall gamma higher than 
unity, due to the cathode-ray tube. A reduction in the gamma may 
be more satisfactorily made with tubes of the dissector type, where 
the noise is negligible in the black portions of the picture, than in a 
tube of the orthicon type, where the noise is determined by the im- 
pedance of the tube and the first amplifier stage. A reduction of 



April, 1942] 



COLOR TELEVISION 



341 



gamma in the latter case results in increased noise in the blacks, if 
all other factors remain equal. 

The introduction of color does not change many of the design re- 
quirements which are ordinarily encountered in monochromatic tele- 




T^ * - I r-WWMri 

I 120 CYCLE SAWTOOTH WAVE ^ _^. ^. 7 ^. ^X- 



0-C MAGNETIC BRAKE 



PHASING BUTTON 

/NORMALLY CLOSED 



DERIVE FROM L.F DISCHARGE TUBE 



BROWNING PULLEY 
IVP30 




BRASS OR OILITE 



STEEL 



FIG. 25. Seven-inch receiver brake circuits and design. 

vision studio equipment. Certain factors, however, are worthy of 
mention. The color-field frequency of 120 per second necessitates 
freedom from 60-cycle hum in the synchronizing generator and scan- 
ning equipment and, to a lesser degree, in video amplifiers. Sixty- 
cycle components present in the synchronizing generator or scanning 
equipment cause loss of interlace and in the video equipment cause 
flicker at a 20-cycle rate resulting from the beat between the 60- 



342 GOLDMARK, DYER, PIORE, AND HOLLYWOOD [j. S. M. p. E. 

cycle hum and the 40-cycle picture components. Hum may be 
eliminated easily by operating the equipment on a 120-cycle power- 
source. 

Good low-frequency response is necessary in video amplifiers to 
pass the 40-cycle picture components properly. The video control 
equipment for color is somewhat more complex than for black-and- 
white transmission, as it seems advantageous to control the gain and 
possibly the background of each color independently, as previously 
mentioned. 




FIG. 26. Seven-inch color television receiver, showing synchronizing 
brake and driving-motor assemblies. 

(Block diagrams of a color television system using dissector and 
orthicon camera tubes are shown in Figs. 12 and 13. Photographs 
of the orthicon direct pick-up color-camera are shown in Figs. 14 
and 15. As previously mentioned, it is essential that the black 
level be established at the camera since manual control of the d-c 
level for each color would seem a tremendous task. This is done by 
applying the blanking pulses to the grid of the orthicon and to the 
cathode of the dissector. 

The video amplifiers in a color television camera channel are con- 
ventional except for the color-mixer. Manual control of gain and 
brightness for each color 'are achieved by the equipment shown in 
Fig. 16. The color-mixer amplifier may be described as an elec- 



April, 1942] 



COLOR TELEVISION 



343 



tronic switch combined with three separate amplifiers, each with its 
own gain and brightness controls. The video signal is switched by 
means of suitable timing pulses (Fig. 17) applied to the screen grids of 
the 6A67 switching amplifier tubes. The pulses are so timed as to 
operate each amplifier in succession, turning on one as another is 




FIG. 27. Generation of filter shapes 



turned off. This switching occurs during the blanking period, and 
switching transients are removed by subsequent clipping of the re- 
combined signals. Blanking is injected on the cathode of each 
switching amplifier tube and the individual brightness of each color 
is adjustable by bias controls. The switching pulses are generated 
by the "ring-frequency divider" circuit shown in Fig. 18. 



344 



GOLDMARK, DYER, PIORE, AND HOLLYWOOD [j. s. M. p. E. 



The non-storage dissector is more easily adapted to a filter disk, 
since it is necessary that the optical image on the cathode be of the 
correct color only at the point which is being scanned. The orthicon, 
on the other hand, being of the storage type, requires that only one 
color be present in the optical image for one complete scanning-field 
period, prior to the actual scansion of a given point. It is possible 
to design a disk to fulfill this requirement, but it will be of consider- 
ably larger diameter than the filter disk for the dissector. Fig. 19 




FIG. 28. Typical receiver filter disk design. 

shows the filter disk for the orthicon, where the contamination of 
colors due to the curvature of the filter segments is not more than 10 
per cent. A filter-drum as shown in Fig. 20 accomplishes the same 
purpose with less contamination and less space. The drum is phased 
so that the shadow of the slotted rods holding the filters follows the 
scanning line. 

Receiving Equipment. Practically any good black-and-white tele- 
vision receiver design may be the basis of a color television receiver. 
Typical color receivers are shown in Figs. 21, 22, 23 and 24. The* 
additional equipment required will be the color disk, with its driving 
and synchronizing means. 



April, 1942] 



COLOR TELEVISION 



345 



Usually additional precautions should be taken to insure complete 
d-c component insertion and freedom from hum. The cathode- 
ray tube should be magnetically shielded to minimize pairing due to 
60-cycle hum. It is fully as important, however, that attention be 
paid to sources of hum in the scanning circuit, such as common ground 
returns carrying alternating current, insufficient power-supply 




FIG. 29. Drum receiver. 

filtering, or magnetic coupling from power equipment into scanning 
transformers. It has sometimes been found economical to inject 
hum in opposing phase to neutralize a-c fields that otherwise may be 
difficult to remove. This is possible only when the interfering hum 
remains constant in phase and amplitude with respect to the neu- 
tralizing components. Another effect which sometimes tends to 
destroy interlacing is found in the electrostatic charges which ac- 
cumulate on the rapidly moving color disk. Variations in the charge 



346 GOLDMARK, DYER, PIORE, AND HOLLYWOOD [j. S. M. p. E. 

over the surface of the disk produce movement of the scanning lines 
as the disk rotates. It is possible to remove the charge with a semi- 




FIG. 30. Rectangular flat-screen cathode-ray tube for 
color television. 



conductive coating on the cathode-ray tube face or with other elec- 
trostatic means of screening. 

Color disks have been made of metal or of transparent plastics 




FIG. 31. A fluorescent-light unit, used in color work. 

such as Lucite, Plexiglass, etc. Wratten filters may be obtained 
coated on a 0.010-inch acetate stock which can be riveted to the 
plastic or metal disk. The disk may be rotated by a synchronous 



April, 1942] 



COLOR TELEVISION 



347 



motor or by an asynchronous motor with auxiliary synchronizing 
means. Owing to lack of synchronism between the power supplies 
of New York, Connecticut, and New Jersey, and also owing to lack 
of standard synchronous motors of 1200-rpm type, it was found 
desirable to drive the disk with inexpensive induction -type motors 
and synchronize it by means of a phonic motor or a magnetic brake. 
Satisfactory phonic motors have been constructed driven by a single 
6V6 tube, but the brake arrangement is preferred (Fig. 25). A 
photograph of the same brake assembly as used on the receiver of 
Fig. 23 is shown in Fig. 26. The 120-cycle voltage is derived from 




FIG. 32. View of a bank of fluorescent lights. 

the low -frequency scanning circuit and is mixed with a similar voltage 
from a small generator on the disk shaft. The sum of the voltages 
is then rectified and the resulting direct current applied to the mag- 
netic brake. A departure in the disk phase with respect to the scan- 
ning produces a corresponding correction on the part of the brake. 

The generation of a properly shaped filter disk is shown in Fig. 27. 
This shape is suitable for a receiving or transmitting tube with short 
decay or storage times. The curve obtained in Fig. 27 is an envelope 
of the position of a scanning line as traced on to the filter which is 
moving with the line. The required filter shape for a given mechani- 
cal arrangement is obtained by developing curves which make allow- 
ance for positive and negative tolerances to take care of fluctuation 
in the disk position, viewing angle, and screen decay (Fig. 28). 



348 GOLDMARK, DYER, PIORE, AND HOLLYWOOD [j. s. M. P. E. 

Generally, the minimum disk diameter is about twice the outside 
diameter of the tube plus one or two inches. The optimal location 
will be determined by such factors as the distance from the disk 
shaft to the picture frame, but can be determined for any particular 
arrangement. 

Color-drums have been used at the receiver as well as at the 
transmitter instead of color-filters. A short cathode-ray tube can 
be placed within the drum (Fig. 29). The drum is designed for a 
lower speed of revolution than is usually possible with the disk. 
Successful drums have been built to operate at a speed of 600 rpm 
or one-half the usual disk speed. 

The small table-model receiver shown in Fig. 23 utilizes a cathode- 
ray tube developed especially for color-pictures. The screen of the 
tube is flat and has the exact shape of the final image (Fig. 30). The 
tube produces a picture equivalent in size to that of a conventional 
7-in. round tube. Thus the color disk is 15 in. in diameter. A 10- 
in. lens with a focal length of 12 in., which is built into the receiver, 
increases the image to correspond to that of a conventional 9-in. 
tube. Owing to low magnification, distortion and decrease of the 
viewing angle are appreciably reduced. 

ACKNOWLEDGMENT 

All members of the Columbia Broadcasting System Television 
Engineering Department have been concerned with this develop- 
ment. In particular, the active cooperation of Messrs. Goetz, 
Freundlich, Harcher, Erde, Doncaster, Reeves, and Haas is gratefully 
acknowledged . 

The CBS Television Program Department has unselfishly con- 
tributed by preparing suitable test and demonstration material. 

Thanks are due to Mr. Adrian Murphy for his constructive criti- 
cism, active encouragement and enthusiastic support throughout the 
entire developmental period. 

REFERENCES 

1 DINSDALE, A. : "Recent Advances in Television: Television by Daylight and 
Television in Colors," Television, I (Aug., 1928), p. 9. 

2 "Television in Colors; Bell Telephone Engineers Transmit and Reproduce 
Scenes in Their Natural Hues," Tel. & Tel. Age (July 16, 1929), p. 315. 

3 MARCHANT, F. W. : "New Baird Colour Television System," Television 
and Short Wave World, XII (Sept., 1939), p. 541. 

4 Commission Internationale de 1'eclairage. Comptes Rendus, des Seances, 
1931, Cambridge Univ. Press (1932), p. 25. 



April, 1942] COLOR TELEVISION 349 

6 HARDY, A. C.: "Handbook of Colorimetry," Cambridge, Mass., M. I. T., 
Technology Press (1936). 

HODGMAN, C. D.: "Handbook of Chemistry and Physics," Cleveland, 
Chemical Rubber Co. (1937). 

6 HARDY, A. C., AND WURZBURG, F. C., JR.: "The Theory of Three-Color 
Reproduction," /. Opt. Soc. Amer., XXVII (July, 1937), p. 227. 

7 LEVERENZ, H. W., AND SEITZ, F. : "Luminescent Materials," /. Applied 
Physics, X (July, 1939), p. 479. 

8 TAYLOR, A. H., AND KERR, G. P.: "The Distribution of Energy in the 
Visible Spectrum of Daylight," /. Opt. Soc. Amer., XXXI (Jan., 1941), p. 3. 

9 PERRY, E. S., Amer. Inst. of Sci., XLIV (1892), p. 192. 

PORTER, T. C.: "Study of 'Flicker,' " Proc. Royal Soc. Lond., LXX (July 29, 
1902), p. 313. 

10 ENGSTROM, E. W. : "A Study of Television Image Characteristics," Proc. 
I. R. E., XXIII (Apr., 1935), p. 295. 



BIBLIOGRAPHY 

Patents 

VON BRONK, O.: No. 155,528 (German), Oct. 22, 1904. Method and device 
for making visible at a distance pictures or objects under temporary decomposi- 
tion of the picture into parallel rows of points. 

BON JAWORSKI, W., AND FRANKENSTEIN, A. : No. 172,376 (German), June 21, 
1906. Method and device for reproducing at a distance pictures and objects by 
means of selenium cells through color-filters, and decomposition of the pictures in 
groups of points by means of mirrors. 

ADAMIAN, J.: No. 7219 (British), 1908. Improvement in electrically con- 
trolled apparatus for seeing at a distance. 

ANDERSEN, A. C., AND ANDERSEN, L. S.: No. 30,188 (British), 1909. Tele- 
vision apparatus. 

ZWORYKIN, V. K. : No. 255,057 (British), July 3, 1926. Television apparatus. 

KELL, R. D.: No. 308,277 (British), Nov. 21, 1929. Television apparatus. 
No. 1,748,883 (U. S.), Feb. 25, 1930. Transmission of pictures. 

SIEMENS-SCHUCKERTWERKE. : No. 387,206 (British), May 30, 1931. Cathode- 
ray tubes. 

VON ARDENNE, M.: No. 388,623 (British), June 8, 1932. Television method. 

HAMMOND, J. H., JR.: No. 1,725,710 (U. S.), Aug. 20, 1929. System and 
method of television. 

IVES, H. E.: No. 390,158 (British), Sept. 28, 1931. Television. 

BAIRD, J. L.: No. 418,527 (British), May 25, 1933. Television. (Noted in 
Television & Short Wave World, 8 (July, 1935), p. 385.) 

LOEWE, D. S., AND SCHLESINGER, K. : No. 426,138 (British), Sept. 29, 1933. 
Cathode-ray tubes. (Noted in Television & Short Wave World, 8 (Aug., 1935), p. 
458.) 

FERNSEH, G. m. b. H.: No. 432,989 (British), March 6, 1934. Television 
receivers and cathode-ray tubes. (Noted in Television & Short Wave World, 8 
(Dec., 1935), p. 715.) 



350 GOLDMARK, DYER, PIORE, AND HOLLYWOOD [j. s. M. P. E. 

No. 434,868 (British), March 6, 1934. Color television and cathode-ray tubes. 
(Noted in Television & Short Wave World, 8 (Jan., 1936), p. 28.) 

HEWITT, W. S.: No. 439,494 (British), June 5, 1934. Television. (Noted 
in Television & Short Wave World, 9 (April, 1936), p. 208.) 

JESTY, L. C. (G. E. Co.): No. 443,896 (British), Oct. 6, 1934. Television. 

BAIRD, J. L.: No. 473,323 (British), May 9, 1936. Television. 

STANDARD TELEPHONES & CABLES, LTD. : No. 500,005 (British), Feb. 18, 1938. 
Cathode-ray tubes. (Noted in Television & Short Wave World, 12 (June, 1939), 
p. 346.) 

CARPMAEL, A.: No. 502,358 (British), Sept. 14, 1937. Television. 

SMYTH, C. N.: No. 508,037 (British), Dec. 24, 1937. Television. (Noted in 
Television & Short Wave World, 12 (Nov., 1939), p. 647.) 

BAIRD, J. L.: No. 508,039 (British), Dec. 24, 1937. Television. (Noted in 
Television & Short Wave World, 12 (Nov., 1939), p. 648.) 

No. 512,855 (British), March 22, 1938. Television. 

Articles and Papers 

DER FERNSEHER DE GEBRUDER ANDERSEN: "Television Apparatus of Ander- 
son and Andersen," Zeits.f. Schwachstrom Technik, IV (1910), No. 17, p. 452. 

HUBBARD, H. D.: "Colored Scenes, Motion, Sound Are Envisaged," Radio 
World. XI (Apr. 23, 1927), p. 13. 

"Motion Pictures in Color Radio's Next," Radio Digest, XX (Mar. 11, 1927), 
p. 13. 

"Colored Television Due," Radio World, XII (Feb. 4, 1928), p. 7. 

DINSDALE, A.: "Recent Advances in Television: Television by Daylight and 
Television in Colors," Television, I (Aug., 1928), p. 9. 

JEVONS, J. C.: "Television in Colour," Popular Wireless, XIII (Aug. 25, 
1928), p. 829. 

TIERNEY, C. : "My Impressions of Daylight and Colour Television," Television, 
I (Aug., 1928), p. 7. 

TILTMAN, R. F. : "Television in Natural Colors Demonstrated," Radio News, 
X (Oct., 1928), p. 320. 

"How Colour Television Images Are Obtained," Television, I (Sept., 1928), p. 
25. 

LARNER, E. T.: "Practical Television," Nature, CXXII (Aug. 18, 1928), p. 
232. 

GLEASON, C. S. : "High Frequencies for Color Television," Radio News, X 
(Jan., 1929), p. 632. 

IVES, H. E.: "Television in Color," Bell Lab. Record, VII (July, 1929), p. 
439; Reprint: Instrument World, II (1929), No. 17, p. 133. 

ANDERSEN, E.: "Ein Vorschlag zum Fernsehen in Naturlichen Farben" (A 
vSuggestion for Television in Natural Color), Fernsehen, I (Apr., 1930), p. 179. 

NEUBERGER, A.: "Farbiges Fernsehen (Color Television)," Fernsehen, I 
(Feb., 1930), p. 62. 

IVES, H. E.: "Television in Color," Television Engr., XXXIII (Aug., 1929), 
No. 8, p. 19. 

"Television in Colors: Bell Telephone Laboratories Engineers Transmit and 



April, 1942] COLOR TELEVISION 351 

Reproduce Scenes in Their Natural Hues," Tel. & Tel. Age (July 16, 1929), No. 
14, p. 315. 
References on same subject: 

Sci. News Letter, XVI (July 6, 1929), p. 3. 
Telephony, XCVII (July 6, 1929), p. 23. 
Wireless World, XXV (July 10, 1929), p. 37. 
Science, LXX (July 5, 1929), supp. p. x. 
Radio World, XV (July 20, 1929), p. 11. 
Elec. Record, XLVI (Aug., 1929), p. 91. 
Pop. Science, CXV (Sept., 1929), p. 25. 
Radio News, XI (Sept., 1929), p. 203. 
Pop. Wireless, XVI (Sept. 14, 1929), p. 11. 
Bell Tel. Quart., VIII (Oct., 1929), p. 344. 
Radio, XI (Oct., 1929), p. 35. 
Nature, CXXTV (Aug. 19, 1929), p. 241. 

"Color Television," Western Elec. News, XVIII (Aug., 1929), p. 18. 

IVES, H. E., AND JOHNSRUD, A. L. : "Television in Colors by a Beam Scanning 
Method," /. Opt. Soc. Amer., XX (Jan., 1930), p. 11. 

"Color Television Reported on Way," Television Engr., XXXIV (Mar., 1930), 
p. 40. 

COOK, S. P.: "On an Electromagnetic Theory of Sight and Color Vision," 
Phys. Rev. (1930), p. 790 (Abstract). 

GRADENWITZ, A.: "Ahronheim's Television in Natural Colors," Television, 
III (June, 1930), p. 177. 

IVES, H. E.: "Television in Color from Motion Picture Film," /. Opt. Soc. 
Amer., XXI (Jan., 1931), p. 2. 

"Photoelectric Cells for Color Television," Television, VII (Jan., 1934), p. 36. 

WILSON, J. C. : "Trichromatic Reproduction in Television," /. Royal Sci. 
Arts, LXXXII (June, 1934), p. 341; Television, VII (June, 1934), p. 268. 

ASHMORE, R. L. : "Color Values in Television," Television & Short Wave 
World, VIII (Sept., 1935), p. 516. (A discussion of the pick-up of colored scenes 
and their reproduction in black and white.) 

"Cheap Screen Invention Reported for Television in Colors with Sound," 
Tel. & Tel. Age (July 1, 1935), p. 146. 

"Television Development Color Broadcasting System," Electrician, CXIV 
(June 14, 1935), p. 805 (News Item). 

"Television in Color Reported in Belgium," Motion Picture Herald, CXIX 
(June 8, 1935), p. 90. 

"Natural Color Television Forecast by New Patent (U. S. 2,055,557)," Sci. 
News Letter, XXX (Nov. 28, 1936), p. 345. 

"Television Pictures Shown in Color," Pop. Sci., CXXXI (July, 1937), p. 53. 

BABITS, V. A.: "Television in Colors," Television & Short Wave World, X 
(Aug., 1937), p. 480. 

"Baird Color Television," Television & Short Wave World, XI (March, 1938), 
p. 151. 

PRAGER, W. L.: "Color in Broadcasting Studied by New Hollywood Tele- 
vision Group," American Cinematographer t XIX (Apr., 1938), p. 160. 



352 GOLDMARK, DYER, PIORE, AND HOLLYWOOD 

"Television in Color Covered by New Patent" (U. S. Pat. 2,109,773), Set. 
News Letter, XXXIV (July 30, 1938), p. 70. 

"Television in Color," Electrician, CXX (Feb. 18, 1938), p. 197. 

OTTERBEIN, G. : "Ein Farblichtrelais" (Colored Light Relay) (Its Use in Color 
Television), T. F. T., XXVII (Spec. Issue) (Nov., 1938), p. 550. 

PRESSLER, H.: "Uber die Eildfeldzerlegung bei der Farbenfernsehubertragun" 
(Resolving the Field of the Image in Color Television Transmission), Ferns ehen 
(supp. to F. T. M.) (Dec., 1938), No. 12, p. 89. 

MAR CHANT, F. W. : "New Baird Color Television System," Television & 
Short Wave World, XII (Sept., 1939), p. 541. 

NEMES, T. DE: "Color Television with Electrical Color Filters," Television & 
Short Wave World, XII (Feb., 1939), p. 73. 

Grande-Bretagne. La Television en Couleurs. (Great Britain. Color 
Television), Rev. Teleph., Teleg., et TSF, XVI (Feb., 1939), p. 169. 

PRESSLER, J.: "La Television en Colores" (Television in Color), Revista 
Telegrafica, XXVII (Mar., 1939), p. 169. (This is a .translation of an article in 
German.) 

MAYBANK, N. W. : "Color Television" (Baird Television system described), 
Wireless World, XLV (Aug. 17, 1939), p. 145. 

REICHEL, W. : "Der Mehrfachzeilenspring" (Multiple Interlacing), Fernsehen, 
I (Aug., 1939), No. 5, p. 171. 

"Television in Natural Color," Communications, XX (June, 1940), p. 8. 

CBS Publicity Release Color Television Achieves Realism, New York Times 
(Sept. 5, 1940). 

ROSENTHAL, A. H. : "The Skiatron A New Scophony Development Towards 
Large Screen Television Projection," Electronics and Television & Short Wave 
World, XIII (Feb., 1940), p. 52; (Mar., 1940), p. 117. 

"A System of Large Screen Television Reception Based on Certain Electron 
Phenomena in Crystals," IRE Proc., XXVIII (May, 1940), p. 211. 

"Color Television Demonstrated by CBS Engineers," Electronics, XIII (Oct., 
1940), p. 32. 



NEW STOP BATH AND FIXING BATH FORMULAS AND 
METHODS FOR THEIR REVIVAL* 



J. I. CRABTREE, L. E. MUEHLER, AND H. D. RUSSELL** 

Summary. Several substitutes for acetic acid are recommended for use in acid 
stop baths and fixing baths including white vinegar, a mixture of sodium acetate and 
sodium acid sulfate, sodium bisulfite, and citric acid. 

Methods of testing stop baths to determine the exhaustion point are described and 
procedures given for their revival by the addition of acid. 

A formula for a new rapid fixing bath is also given, including instructions for re- 
viving fixing baths by the addition of acid at intervals. 

In view of the requisition of many chemicals for defense purposes 
it may be impossible to secure in all parts of the country some es- 
sential photographic chemicals. An investigation has therefore been 
made with a view to making available a choice of chemicals to replace 
those ordinarily used in compounding stop baths and fixing baths. 

STOP BATHS 

Tests have indicated that for acid stop bath purposes a variety 
of acidic compounds may be substituted for the acetic acid usually 
employed, namely, a mixture of sodium acid sulfate or sulfuric acid 
and sodium acetate, citric acid, sodium bisulfite or potassium meta- 
bisulfite, and white vinegar. 

Sodium Acid Sulfate and Sodium Acetate. An almost identical sub- 
stitute for the Kodak SB-1 acid stop bath can be compounded from 
sodium acetate and an acidic substance such as sodium acid sulfate 
or sulfuric acid as follows: 

Kodak SB-8 Stop Bath 





Avoirdupois 


Metric 


Water 


64 ounces 


500 cc 


Kodak Sodium Acetate (desiccated) 


2 oz 290 grains 


20 grams 


Kodak Sodium Acid Sulfatef 


5 l /4 ounces 


40 grams 


Water to make 


1 gallon 


1 liter 



* Received January 14, 1942. 
** Kodak Research Laboratories, Rochester, New York. 

353 



354 CRABTREE, MUEHLER, AND RUSSELL [j. s. M. p. E. 

f OR Kodak Sulfuric Acid, 5% (22'/2 fluid oz per gallon; 175 cc per liter). 
To make 5 per cent sulfuric acid, add slowly 1 part by volume of Kodak Sulfuric 
Acid, C. P., to 19 parts by volume of cold water and mix carefully with stirring. 
The acid must be added slowly to the water and not the water to the acid, otherwise 
the solution may boil with explosive violence. 

In the case of highly alkaline developers a somewhat more concentrated acid 
stop bath is necessary such as Kodak SB-la which should preferably be used 
only in conjunction with Kodalk or caustic developers in order to avoid blistering. 
The stop bath should likewise be followed by a short water rinse. The formulas 
Kodak SB-8a, SB-7a, and SB-6a below are the more concentrated stop baths 
corresponding to SB-8, SB-7, and SB-6. 

Kodak SB-8a Stop Bath 

Avoirdupois Metric 

Water 64 ounces 500 cc 

Kodak Sodium Acetate (desiccated) 6 3 / 4 ounces 50 grams 

Kodak Sodium Acid Sulfate* 13V4 ounces 100 grams 

Water to make 1 gallon 1 liter 

* OR Kodak Sulfuric Acid, 5% (56 fluid oz per gallon; 440 cc per liter). 
To make 5 per cent sulfuric acid, add slowly 1 part by volume of Kodak Sulfuric 
Acid, C. P., to 19 parts by volume of cold water and mix carefully with stirring. 
The acid must be added slowly to the water and not the water to the acid, other- 
wise the solution may boil with explosive violence. 

Citric Acid. A 1.5 per cent solution of citric acid constitutes a very 
satisfactory acid stop bath although, if an excess is carried over into 
the fixing bath, it tends to impair the hardening properties of the 
fixing bath. 

The Kodak Testing Solution A for Stop Baths should be used to 
determine when the bath is exhausted. When exhausted, the bath 
should be discarded and a fresh bath prepared. 

Suitable formulas are as follows : 

Kodak SB-7 Stop Bath 

Avoirdupois Metric 

Kodak Citric Acid 2 ounces 15 grams 

Water to make 1 gallon 1 liter 

Kodak SB-7a Stop Bath 

Avoirdupois Metric 

Kodak Citric Acid 5 ounces 37.5 grams 

Water to make 1 gallon 1 . liter 

Sodium Bisulfite. As a substitute for the Kodak SB-1 acid stop 
bath, the formula Kodak SB-6 may be used, namely, 4 ounces per 



April, 1942] STOP AND FIXING BATH FORMULAS 355 

gallon of Kodak Sodium Bisulfite. For Kodak SB-la the proportion 
is 10 ounces of bisulfite per gallon (Kodak SB-6a) as follows: 

Kodak SB-6 Stop Bath 

Avoirdupois Metric 

Kodak Sodium Bisulfite 4 ounces 30 grams 

Water to make 1 gallon 1 liter 

Kodak SB-6a Stop Bath 

Avoirdupois Metric 

Kodak Sodium Bisulfite 10 ounces 75 grams 

Water to make 1 gallon 1 liter 

White Vinegar. White vinegar or distilled vinegar is a dilute solu- 
tion of acetic acid, its strength varying from 3 to 15 per cent. The 
white vinegar sold in grocery stores in 1-gallon jugs is usually of 4.5 
per cent acidity. The strength of vinegar is also designated by its 
"grain" content, 1 per cent acidity being approximately equal to 10- 
grain; therefore, a 45-grain vinegar represents a vinegar of 4.5 per 
cent acidity. Distilled vinegar containing 4 to 15 per cent acidity 
(40- to 150-grain) may be purchased from vinegar manufacturers or 
wholesale food supply houses in barrels. 

For a substitute Kodak SB-1 stop bath, dilute 1 part of white 
vinegar (4.5% = 45-grain) with 2 parts of water. As a substitute 
for the Kodak SB-la, use white vinegar full strength. 

Brown vinegar may be used instead of white vinegar but it is not 
as satisfactory. The odor is offensive and it tends to cause excessive 
foaming of the stop bath and fixing bath. This tendency to foam 
may, however, be decreased by the use of Kodak Anti-Foam Solu- 
tion. 

It is necessary that the acid stop bath remain acid throughout its 
life and the acidity can be checked by the use of the Kodak Testing 
Solution A for Stop Baths which may be used in either of two ways: 
(1) a small quantity of the stop bath is withdrawn and placed in a vial 
and the testing solution added to this; or (2) the indicator solution 
is added to the tray or tank containing the stop bath in the proportion 
indicated by the following table. When the bath turns purple, this 
is a signal that the bath should be discarded or revived. 

All stop baths containing acetic acid, vinegar, sodium bisulfite 
(potassium metabisulfite), or sodium acetate and sodium acid sulfate 
may be renewed or revived with acid after the bath containing the 
indicator turns purple. The acid used in the revival may be any 



356 CRABTREE, MUEHLER, AND RUSSELL [J. S. M. P. E. 

Concentration of Kodak Testing Solution A for Stop Baths 
TRAY 



1 Quart 
or 
1 Liter 


2 Quarts 
or 
2 Liters 


1 Gallon 
or 
4 Liters 


Ice 


2cc 


4 CC 


15 minims 


30 minims 


60 minims 


l /4 dram 


l /z dram 


1 dram 




V4 vial 


Vt vial 



Vs ounce 
TANK 

(White Card or Tray on Bottom of Tank) * 
1 Gallon 3V2 Gallons 

Vz dram 2 drams 

: /4 vial 1 /4 ounce 

1 vial 

* For larger tanks immerse the white tray or card to a depth of about 10 inches. 

acidic compound which does not impair the properties of the bath. 
The common acids which are satisfactory include acetic acid, hydro- 
chloric acid, sulfuric acid, and sodium acid sulfate. Boric acid and 
sodium bisulfite are not strong enough acids to give the proper re- 
action with the indicator solution unless they are added in excessive 
quantities. Citric acid, tartaric acid, and other hydroxy organic 
acids may be used in small quantities, otherwise, when the stop bath 
is carried into the fixing bath, the hardening properties of the fixing 
bath may be impaired. 

Acid stop baths containing sodium bisulfite (potassium metabi- 
sulfite) need to be revived more frequently than the others because 
a 3 per cent solution of sodium bisulfite (potassium metabisulfite) has 
only one-third the developer life as indicated by the Kodak Testing 
Solution A for Stop Baths in comparison with the SB-1 acetic acid 
stop bath. 

REVIVAL OF STOP BATHS 

The substitute acid stop baths can be revived in the following man- 
ner but they should be discarded whenever they become discolored or 
accumulate suspended matter: 

(1) Trays and Small Tanks. Add 20 drops of Kodak Testing 
Solution A for Stop Baths to each quart (1 liter) of rinse bath. The 
indicator is yellow in the acid bath but turns^ purple when the acid 
has been neutralized If the indicator turns purple, the stop bath 



April, 1942] STOP AND FIXING BATH FORMULAS 357 

may be revived by the addition of either the following formula, 
Kodak SB-8R, or a 5 per cent solution of sulfuric acid* with stirring. 
Add the acid slowly until the bath turns a light yellow color but do 
not add an excess of the acid. 

Kodak SB-8R 

Avoirdupois Metric 

Kodak Sodium Acid Sulfate 3 l / t ounces 97 . 5 grams 

Water to make 32 ounces 1 . liter 

(2) Deep Tanks. Add 4 drams (15 cc.) of Kodak Testing Solution 
A for Stop Baths to each 10 gallons (40 liters) of the bath and place 
either a white card or a white tray in the bath such that the tray or 
card is immersed for a distance of about 10 inches to serve as a back- 
ground for viewing the color of the bath. If the bath requires 
revival with acid, the solution will appear reddish purple in tungsten 
light. Add either a 5 per cent solution of sulfuric acid or the sodium 
acid sulfate revival bath, Kodak SB-8R, at the rate of 1 ounce per 
gallon until the purple color changes to light yellow, mixing the solu- 
tion completely after each addition of acid. The indicator, after 
addition to the bath, may fade with time so that a fresh quantity 
should be added before each revival. 

The Kodak Testing Solution A for Stop Baths may stain prints or 
films if they are left in the bath longer than 2 minutes. In this case 
it is advisable to test only a small portion of the bath with the test 
solution during revival of the bath with acid. In the test, fill the 
vial from the Kodak Testing Outfit within one-quarter inch of the 
top with the stop bath to be tested and add 1 drop of the Testing 
Solution A for Stop Baths ; mix the solution by placing a finger over 
the open end of the vial and shaking the vial. If the solution turns 
purple, add Kodak SB-8R at the rate of 1 ounce per gallon and test 
the bath after each addition of acid. Continue adding acid to the 
bath (in the tank or tray) until a light yellow color is obtained, when 
1 drop of test solution is added to the sample in the vial. Do not 
add an excess of acid. 

The life of the regular acetic acid stop baths such as Kodak SB-1, 
Kodak SB-la, and Kodak SB-5 may also be prolonged by the ad- 

* To make 5 per cent sulfuric acid, add slowly 1 part by volume of Kodak Sul- 
furic Acid, C. P., to 19 parts by volume of cold water and mix carefully with stir- 
ring. The acid must be added slowly to the water and not the water to the acid, 
otherwise the solution may boil with explosive violence. 



358 CRABTREE, MUEHLER, AND RUSSELL fj. S. M. P. E. 

dition of Kodak SB-8R, according to the above procedure. The 
Kodak SB-5 Stop Bath containing acetic acid and an anti-swelling 
agent (sodium sulfate) is recommended exclusively for use with film. 
When used at normal temperatures, the bath may be revived with 
acid at frequent intervals but the sulfate becomes diluted, which im- 
pairs the usefulness of the bath at high temperatures. When the 
bath is used at high temperatures, it tends to become discolored and 
accumulates a sludge more readily which, unless filtered or siphoned 
off, may adhere to the film. It is desirable, therefore, to renew the 
bath frequently. Other stop baths should be discarded after they 
have been revived 3 or 4 times or when they become discolored. 

Formulas for 1 -liter, 1 -gallon, 17-gallon, and 48-gallon quantities 
of the Kodak SB-6, SB -7, and SB-8 stop baths are given in Table I. 

Substitutes for Non-Swelling Stop Bath SB-5. In compounding a 
substitute for the Kodak non-swelling acid stop bath SB-5 add to any 
of the above stop bath formulas sodium sulfate (desiccated) in the 
proportions given below : 

Quantity of Sodium 

Volume of Sulfate (Desiccated) 

Stop Bath Required 

1 liter 45 grams 

1 gallon 6 ounces 

17 gallons 6 Ib 6 ounces 

48 gallons 18 Ib 

Add the desiccated sodium sulfate slowly with stirring, otherwise 
the sulfate will cake and then dissolve with difficulty. 

Revival of Chrome Alum Stop Baths. The chrome alum stop baths 
Kodak SB-3 and Kodak SB-4 can not be revived by the above pro- 
cedure using the Kodak Testing Solution A for Stop Baths. The re- 
vival of a chrome alum bath is more complicated than that in the case 
of an acetic acid bath, because of several peculiar properties of chrome 
alum solutions. The >H value of a freshly prepared chrome alum 
solution is approximately 3.0 and the maximum hardening properties 
in the fresh solution are obtained between pH values of 3.0 and 4.0. 
With slight use when a small quantity of developer is carried into 
the stop bath, the pH value increases but, on standing, the solution 
undergoes a change in an attempt to adjust itself to its original pH. 
value of 3.0. This change involves the formation of basic chromium 
complexes with the liberation of sulfuric acid. The presence of 
basic chromium complexes increases the sludging and scumming 



April, 1942] 



STOP AND FIXING BATH FORMULAS 



359 






% 

S I 



H 5 






W 



So a 

s- 



II 

a a 



CO CO 

d ^O 

II 



$ d e 
a ^ BQ 



^ 
i-t IN 



"83 



co co 

11 



CO 00 



co 

ll 



CO V) 

5 



co co co 

111 



a H 



as 



a -g 3 

00 CO 00 S 'S 

" li 

d 2 

QJ CJ 

a v- 



CO CO CO CO 

d *O *T3 C3 

o d d o 



Cl O *> 






C, S ctf C 
| O bO O 



$ G 

1 II 



I s 

* 



3 



a % 

O tf 




to g 

SI -a 



s t; 



&i| 

o | o 
J5 > S 



jd fri ^3 

1^5 
Hi 

. o . 

^JH t 
10 <H o 



^j 7g 





II 
11 

** 

+* co 

II 



360 CRABTREE, MUEHLER, AND RUSSELL [J. S. M. P. E. 

tendency of the bath and, if the bath is used only occasionally, the 
bath may suddenly sludge or produce an undesirable scum on the 
film. This condition of the bath can not be readily determined since 
the pH value of the solution may be similar to that of the fresh bath 
and revival with acid will not restore it to its original condition. 
Therefore, it is recommended that if a chrome alum stop bath is 
to be used at infrequent intervals, only a fresh bath should be used. 

Chrome alum baths which are in continuous use may be satis- 
factorily revived if the rate of exhaustion is such that they would 
normally be exhausted within a 24-hour period. In this case the 
condition of the bath is best determined by the use of a glass-elec- 
trode pH meter and the bath should be revived at frequent intervals 
to a pH. value of 3.0 by the addition of a 5 per cent solution of sulfuric 
acid or Kodak SB-8R. 

The condition of the bath may also be determined by titrating a 
25-cc sample of the bath (in 50 cc of distilled water containing 5 cc 
of a 1 : 2500 solution of bromophenol blue indicator) with a solution of 
sulfuric acid (2.5%) or Kodak SB-8R until the solution is just acid 
to the indicator. The indicator solution is blue at a p~H. value of 4.0 
and yellow at a value of 3.0, but the color is somewhat obscured by 
the green color of the chrome alum solution. The color in an acid 
chrome alum solution will be yellowish green and, in an alkaline solu- 
tion, a dirty blue. 

From the quantity of acid required in the titration for 25 cc of 
the stop bath, the quantity for the entire bath can be calculated. 
One cubic centimeter of a 2.5 per cent solution of sulfuric acid per 
25 cc of the bath will require the addition of 2.5 ounces per gallon 
of 5 per cent sulfuric acid. One cubic centimeter of Kodak SB-8R 
per 25 cc of the bath will require the addition of 5 ounces of Kodak 
SB-8R per gallon. 

The above test may be modified as follows to permit the use of 
methyl orange instead of the bromophenol blue indicator if the latter 
is not readily available: Dilute a 25-cc sample of the used SB-3 
stop bath to 250 cc with water and then add 15 cc of a solution of 
methyl orange indicator prepared by dissolving 1 gram in 2500 cc 
of water. If the bath requires replenishment the test solution will 
be lemon-yellow. Titrate as described above with acid. An orange- 
pink or peach color indicates the correct degree of revival but, if too 
much acid is added, the color will be rose-pink. 

Testing the Strength of Acetic Acid Solutions. Acetic acid is sold 



April, 1942] STOP AND FIXING BATH FORMULAS 361 

in a variety of strengths and, in case the precise strength is unknown 
it can be determined approximately by the following procedure: 

Prepare a standard solution of Kodalk by dissolving 3 ounces (85 
grams) in 20 ounces (600 cc) of water. Then carefully add water to 
make 26 ounces (770 cc), stir to obtain complete uniformity, and 
place in a closed quart bottle. 

When measuring liquids, the reading of the graduate should be 
taken from the lowest part of the curved surface of the liquid (menis- 
cus) with the eye held at the same level. 

In a clean 1 -ounce graduate standing on a level surface carefully 
measure exactly 0.5 ounce of the acetic acid solution, the percentage 
strength of which is unknown. Add this to a previously washed 16- 
or 32-ounce graduate standing on a white surface, for example, in a 
white enameled tray. Add 3 to 4 ounces of water, stir with a clean 
stirring rod and follow with 5 drops of the Kodak Testing Solution 
A for Stop Baths. The purple indicator will change to yellow in the 
acid solution. 

Carefully measure 12 ounces of the Kodalk solution into a 16- 
ounce graduate standing on a level surface. Slowly add the Kodalk 
solution (avoid spillage) to the acetic acid solution containing the 
indicator while stirring with a clean stirring rod, and carefully observe 
the color. When nearing the neutralization point, the color will 
change to a dirty yellow, then to a gray, and finally to a distinct 
permanent red-violet. When just sufficient Kodalk solution has been 
added to cause the color to change to red- violet, set the graduate on 
a level surface and read the remaining volume in ounces. Subtract 
this number of ounces from the 12 ounces originally taken to deter- 
mine the number of ounces (including fraction) used to neutralize the 
acid. Multiply the number of ounces used by 10 to give the per- 
centage strength of the acetic acid. 

As an example, the original volume of Kodalk solution was 12 
ounces and the final volume 4 ounces. Therefore, 12 ounces minus 
4 ounces = 8 ounces of standard Kodalk solution used. Acetic 
acid strength = 8 X 10 = 80 per cent. 

The accuracy of the determination depends principally upon the 
care used. For medium and high percentages of acid the method 
has been found to give results to within about 5 per cent of the acetic 
acid strength. 

For lower-strength acid (5 to 30 per cent) better accuracy is ob- 
tained when the starting volume of the Kodalk solution is 4 ounces 



362 CRABTREE, MUEHLER, AND RUSSELL [j. S. M. p. E. 

instead of 12. In this case subtract the remaining volume of the 
Kodalk solution in ounces from 4 to give the volume used. For 
example : 

Original volume of Kodalk solution 4 ounces 

Final volume of Kodalk solution 2 ounces 

Volume of Kodalk solution used 2 ounces 

Acetic acid strength = 2 X 10 = 20 per cent 

For best accuracy with acid strengths of the order of 5 per cent or 
less, use 1 ounce of the Kodalk solution in a 1 -ounce graduate. 
Determine the fraction of an ounce used and multiply this by 10. 
For example: 

1 minus J /4 = 3 A ounce of standard Kodalk solution used, and 
3 /4 X 10 = 7*/2 per cent acetic acid 

FIXING BATHS 

The desirable hardening and anti-sludging properties of acid- 
hardening fixing baths containing acetic and boric acids can not be 
duplicated by other readily available acids. Propionic acid is a 
suitable substitute for acetic acid but it is available commercially 
in only small quantities. 

However, acetic acid can be replaced satisfactorily by sodium 
acetate in combination with another acidic compound such as sodium 
bisulfite or sodium acid sulfate. White vinegar can also be used as 
a substitute for acetic acid in formulas Kodak F-l, F-5, F-6, and 
F-10. 

The hardening and fixing-bath formulas Kodak F-l and F-5 which 
are most commonly used require normally 6 ounces of 28 per cent 
acetic acid per gallon. In these formulas 40 ounces of white vinegar 
should be substituted for the 6 ounces of 28 per cent acetic acid. 

The acid-hardening stock solutions Kodak F-la ( V) and F-5a ( V) 
are given in Tables II and III. Important: These hardening for- 
mulas must be added to a 48 per cent solution of hypo which is 
roughly 4 pounds of hypo contained in 1 gallon of solution. 

The baths Kodak F-3, F-3a, F-4, and F-4a, containing sodium ace- 
tate, together with a suitable acid compound, such as sodium bisulfite 
or sodium acid sulfate, have properties very similar to those of F-5. 

A formula having properties similar to those of the odorless bath 
Kodak F-6 can be compounded by substituting 2 parts of Kodalk 
for 1 part of boric acid in Kodak F-5 (V) or Kodak F-3 described 



April, 1942] 



STOP AND FIXING BATH FORMULAS 



363 



TABLE II 



TO MAKE 

Water (about 125F) 

(50C) 
Kodak Sodium Thiosul- 

fate (Hypo) 
Kodak Sodium Sulfite 

(desiccated) 

Vinegar (4.5%) 45-grain 
Kodak Potassium Alum 
Water to make 



TO MAKE 

Water (about 125F) 

(50C) 
Kodak Sodium Sulfite 

(desiccated) 

Vinegar (4.5%) 45-grain 
Kodak Potassium Alum 
Water to make 



Kodak Substitute Fixing Bath 
Kodak F-l (V} 



1 Gallon 


Avoirdupois 
45 Gallons 


48 Gallons 


Metric 

1 Liter 


64 ounces 


22 x /2 gallons 


24 gallons 


500 cc 


2 pounds 


90 pounds 


96 pounds 


240 grams 


2 ounces 
40 fluid oz 
2 ounces 
1 gallon 


5 3 / 4 pounds 
14 gallons 
5 3 / 4 pounds 
45 gallons 


6 pounds 
15 gallons 
6 pounds 
48 gallons 


15 grams 
310 cc 
15 grams 
1 liter 


Kodak Substitute 


Stock Solution 






Kodak F-la (F) 




Avoirdupois 




Metric 


1 Gallon 


22 l /2 Gallons 


24 Gallons 


1 Liter 


32 ounces 


5*/2 gallons 


6 gallons 


250 cc 


4 ounces 
80 fluid oz 
4 ounces 


5 3 / 4 pounds 
14 gallons 
5 3 A pounds 


6 pounds 
15 gallons 
6 pounds 


30 grams 
620 cc 
30 grams 



1 gallon 22 J /2 gallons 24 gallons 1 liter 



To prepare Kodak F-l ( V] from F-la ( V} add 1 part of cool stock hardener solu- 
tion F-la ( V) slowly to 1 part of cool 48 per cent hypo solution while stirring the 
hypo rapidly, or use the quantities indicated as follows : 



Kodak Substitute Fixing Bath 
Kodak F-l (V) 

Avoirdupois 
1 Gallon 



Metric 

1 Liter 



45 Gallons 48 Gallons 
ounces 15 gallons 16 gallons 330.0 cc 



TO MAKE 

Water (about 125F) 
(50C) 42 

Kodak Sodium Thiosul- 

fate (Hypo) 2 pounds 90 pounds 96 pounds 240.0 grams 

When the hypo is dissolved, add the following quantity of Kodak F-la (V} 

stock hardener: 

Hardener F-la (V) Vz gallon 22 J /2 gallons 24 gallons 500. cc 

Add water to make 1 gallon 45 gallons 48 gallons 1 . liter 



364 



CRABTREE, MUEHLER, AND RUSSELL [J. S. M. P. E. 



below. Formula Kodak F-6 is to be preferred to Kodak F-5 for use 
in trays because the odor of sulfur dioxide is very much less. Kodak 
F-6 is recommended especially for papers but may also be used with 

TABLE m 

Kodak Substitute Fixing Bath 

Kodak F-5 (V) 

TO MAKE 

Water (about 125F) 

(50C) 
Kodak Sodium Thiosul- 

fate (Hypo) 
Kodak Sodium Sulfite 

(desiccated) 
*Kodak Boric Acid, 

crystals 

Vinegar (4.5%) 45-grain 
Kodak Potassium Alum 
Water to make 

* Use crystalline boric acid as specified ; avoid use of the powdered variety which 
dissolves slowly. 

Kodak Substitute Stock Solution 
Kodak F-5a (V) 

Avoirdupois Metric 

1 Gallon 22 l /2 Gallons 24 Gallons 1 Liter 



Avoirdupois 


Metric 


1 


Gallon 


45 Gallons 


48 


Gallons 


1 


Liter 


64 


ounces 


22V2 


gallons 


24 


gallons 


500. 





cc 


2 


pounds 


90 


pounds 


96 


pounds 


240, 





grams 


2 


ounces 


5 3 / 4 


pounds 


6 


pounds 


15. 





grams 


1 


ounce 


2 3 A 


pounds 


3 


pounds 


7.5 


grams 


40 


fluid oz 


14 


gallons 


15 


gallons 


310. 





cc 


2 


ounces 


53/ 4 


pounds 


6 


pounds 


15. 





grams 


1 


gallon 


45 


gallons 


48 


gallons 


1. 





liter 



TO MAKE 

Water (about 125F) 

(50C) 
Kodak Sodium Sulfite 

(desiccated) 

Vinegar (4.5%) 45-grain 
*Kodak Boric Acid, 

crystals 

Kodak Potassium Alum 
Water to make 



32 ounces 5 1 /* gallons 6 gallons 250 cc 



4 ounces 
80 fluid oz 

2 ounces 
4 ounces 
1 gallon 



5 3 /4 pounds 6 pounds 
14 gallons 15 gallons 



2 3 /4 pounds 
5 3 / 4 pounds 



3 pounds 
6 pounds 



22 V2 gallons 24 gallons 



30 grams 
620 cc 

15 grams 

30 grams 

1 liter 



* Use crystalline boric acid as specified; avoid use of the powdered variety 
which dissolves slowly. 

To prepare Kodak F-5 (V) from F-5a (V) add 1 part of cool stock hardener 
solution slowly to 1 part of cool 48 per cent hypo solution while stirring the hypo 
rapidly, or use the quantities indicated in Table IV. 

films if used in combination with an acid stop bath. The acid life of 
Kodak F-6 is approximately only one-half that of Kodak F-5 but, 
by using an intermediate acid stop bath, the life may be extended 



April, 1942] STOP AND FIXING BATH FORMULAS 365 

greatly. The formulas Kodak F-3, F-3a, F-4, and F-4a are given 
in Tables V and VI. 

A New Rapid Fixing Bath. A fixing bath (Kodak F-7) having a 
greater fixing capacity than Kodak F-5 has been compounded by the 
use of ammonium chloride in combination with a relatively high 
concentration of hypo, together with the hardener constituents of 
Kodak F-5. Kodak F-8 is a hypo solution containing ammonium 
chloride which can be used in combination with hardeners Kodak 
F-5a and the substitutes for Kodak F-5a, namely, Kodak F-3a and 
F-4a, to produce a fixing bath having properties similar to those of 
Kodak F-7. Kodak F-8 is not sufficiently concentrated to be used 
with F-5a (V) but a fixing bath containing vinegar can be com- 

TABLE IV 

Kodak Substitute Deep Tank Fixing Bath 
Kodak F-5 (V) 

Avoirdupois Metric 

TO MAKE 1 Gallon 45 Gallons 48 Gallons 1 Liter 

Water (about 125F) 

(50C) 42 ounces 15 gallons 16 gallons 330 cc 

Kodak Sodium Thiosul- 

fate (Hypo) 2 pounds 90 pounds 96 pounds 240 grams 
When the hypo is dissolved, add the following quantity of Kodak F-5a (V) 

Stock Hardener: 

Hardener F-5a (V) 1 / 2 gallon 22 x /2 gallons 24 gallons 500 cc 

Add water to make 1 gallon 45 gallons 48 gallons 1 liter 

pounded similar to formula F-7 by adding the hypo, ammonium chlo- 
ride, and sodium sulfite to one-half the usual quantity of water, 
adding the vinegar before they are all dissolved, then adding the 
boric acid and alum, diluting to volume with water and stirring until 
dissolved. 

When compounding the formulas, the ammonium chloride should 
be added to the hypo solution and not to the final fixing solution 
containing the hardener, otherwise a sludge may form. 

The formula Kodak F-7 (Table VII) is recommended especially 
for use in the machine processing of negative films. It can also be 
used for papers but has no advantage over other formulas which do 
not contain ammonium chloride. With papers it should invariably 
be used in conjunction with an acid stop bath, otherwise, if the bath 
becomes alkaline, dichroic fog is apt to be produced. 



366 



CRABTREE, MUEHLER, AND RUSSELL [j. s. M. p. E. 



en en en 

ess 



. 

*j t-looo^oooo 

'""' 





en 

=3 N N 

^ O 

w> a <N co 
>^^>> 

CD CD CO 00 CD CO O 



-s-s 

c o 

I s 

a a 



o 



III 



a a c a 

3 3 3 3 3 
OOOOO 



o 




April, 1942] 



STOP AND FIXING BATH FORMULAS 



367 



en en en en en 

S | S I fc 

'C ^ o bO bO bi) bo bi 5 

>S i-loooioooo 



3 3.3 -3 -8 -3 -3 a 

^oSccccao 

lilliall 

OOTfiCOOCOOOCOOO 

-^ C<J 05 l-l TJH 



tn tn tn en en 



^ g 

ctf bccua 



<L 


g 


M 



be 


en 
> 


en 

1 
bC 


en 

5 


1 




1 


^ 

1 


(-1 

o 

I 


hJ 


O 


O 


O 








o 




S 


ti 




T 1 




10 


|> 





>0 


TH 




a 


'o 




ri 






H 




1 1 











>> 





















1 


"-M 

c 























rt 

s. 























D^ 


f\ J 


10 Gallons 


I 

1 

CO 


S3 
O 

CO 
O 

T 1 


N 

CM 

jO 


si 

O 

CO 

jQ 


en 

T3 
S 

! 

CO 


en 


1 

O 
i i 


h 

1 


1 
^ 

15 
^ 

c 


^ 

4-) 

s 
s 

*-> 

g 


taining Roch< 


















o 


o 


c 
















en 


tj 





8 
















^> 


-3 


3 


en 

,3 




















^cu" 


-M 


1 


en 

G 


S3 

O 

CO 


N 

O 




N 
O 

CO 


s 


en 

*Ej 


'-5 
43 


^5 


^2 
J3 
3 


S 

0) 

45 


o3 


8, 


~* 


f) 


^ 


^ 


J 


o 


-j 


> 




O 


be 


^H 


^ 


i H 




bo 




C 


rt 


*+H 


iO 


CO 


iO 


-H 


^ 


CO 


W3 


t 


8 


'o 


O 

<u 
















>^ 


VH 

OJ 


c 


<4-c 



2 I 



. 






S r^ 

H ^ 



.2 

ON3N3rt 
CX O O O O bfl 



7 i 







I 



Tt< 



(N CO rH 



(M H 



en tn en en en -. 

8 1 i II 

I I 



a c s^ 



6 
5 



O 1-1 







I 

en *Lo 

3s 
^ b^ s 

S " ^ 



fe ^ : r> 



Bis 
d 
e 



3 t! ^ g 



i- - ^ s 4, 

t I 'S i 11 

i i j f i 

h> -9 o PQ o o _ 

3 > & _J o5 ft 2 

s u ^ -a ^ .* t; 

^ cu c3 T3 rt rt <u 

o * "g 5*8 "85 
g^ w^ w w^ 



368 



CRABTREE, MUEHLER, AND RUSSELL [j. s. M. P. E. 



en en en 

III 

Ui i- i-< CJ 

be bo bo o 



si 



bo ^ 
OOOOOiOOO 

SiO* 00 N- iO i-J 
rH T^i i I 



tn en ^ 

s s 1 

Jn , ^ f 3 -^ . 

. 8 & fe 6 ga 

-5 O C) O C CD O 

^^ CO *^5 ^ ^^ 




G rt 

IS 



I *M 



S *o 



3 C 



* The hardener constituents 
** Use crystalline boric acid 
The following stock hypo sol 



en en en 

y, G *O *O 

III! 

a w o. a 



en J2 

I 

iO > 



a 




aa 



10 o 

CO >-H 



M 



en J2 en 

8? S 
CSC 
3O3 

o a o 



CO CO 



o 
a 





41 

3 J3 

S o 

a e 



W o 

w-S 



5 "8*8 
^ w w 



en en 

II 

a 8, 



Harden 
Add wa 



April, 1942] 



STOP AND FIXING BATH FORMULAS 



369 



.At _!_** 



CO 



<S to to to to to to to 

i o o o o o o o 

8883888 

<N i-l TH i-l TH TH TH 



CO 

^ to f T i r r i T T I r T i T T I f T i 

rJ^ooooooo 

' 28i8S88 

Cq Cq TH TH TH rH TH 



8 

^ 

.s 



to to to to to to to 
ooooo 

c^ioooooo 

TH rH 00 ^H O O O 
(M C<J ^H rH rH rH rH 

A 



CO 

to to to 



<M cq o o o 

rH rH i I O CO 

<M (M C<J (N rH 

A A 





^ 
.3 



g 

to to to to to to to 

o o o o o o o 

cq cq o o o o o 

rH rH rH O> CO O O 

Cq Cq Cq rH i I rH rH 

A A 



I - I 

io tototototototo -* 

2 ^rJ 0000000 C^ 

K^^ cqcqcqoooo rt 

fed rHrHrHiH-^OO 

cqcqcqcqi-HrHTH ' 

A A A .* 



sgi^ 



c^ 



370 CRABTREE, MUEHLER, AND RUSSELL [J. S. M. p. E. 

The advantages of this formula are that (1) the time to clear most 
negative films is less than that for Kodak F-5, and that (2) the 
fixation life is approximately 50 per cent greater than that of Kodak 
F-5. 

A comparison of the hardening properties* of the fixing baths 
Kodak F-3, F-4, and F-5 with and without increased hypo plus 5 
per cent ammonium chloride is given in Table VIII. The results in- 
dicate that the addition of ammonium chloride decreases the harden- 
ing properties slightly, that is, the melting point in most cases is 
10 to 30F lower than in the bath containing only 24 per cent 
hypo. The sludge life was not affected and was equivalent to 80 
per cent of MQ-25 developer.** 

The effect of exhaustion on the time to clear Panatomic-X Roll 
Film was determined in F-5 and in F-5 containing additional hypo 
and ammonium chloride (Kodak F-7). Throughout the exhaustion 
Panatomic-J\T Roll Film was developed in DK-60a, rinsed in SB-5, 
and fixed. The times to clear after 0, 40, 80, and 120 rolls (area of 
roll approx. 80 sq in.) per gallon had been processed in F-5 were 
5 1 /4, 711, and > 30 minutes, respectively, while the times to clear 
in F-7 were l 1 /^, 2 1 /4, 5, and 8 L /% minutes, respectively, as shown in 
Table IX. 

TABLE IX 

Effect of Exhaustion on the Time to Clear Panatomic-X NC Film at70F 

Degree of 
Exhaustion 
(No. of Rolls 
per Gallon) 

Kodak F-5 Kodak F-7 

5*/4 min lYz min 

40 7 min 2 J /4 min 

80 11 min 5 min 

120 > 30 min 8Y 2 min 

An advantage similar to that given by Kodak F-7 is obtained when Kodak F-8 
is used in combination with hardeners Kodak F-3a and F-4. 

* The degree of hardening was determined in a manner similar to that described 
in the paper, "Some Properties of Fixing Baths," by J. I. Crabtree and H. A. 
Hartt, Trans. Soc. Mot. Pict. Eng., XIII, No. 38 (1929), p. 370. 

**MQ-25 

Elon 1 . gram 

Hydroquinone 4 . grams 

Sodium sulfite, desiccated 75 . grams 

Sodium carbonate, desiccated 25.0 grams 

Potassium bromide 5.0 grams 

Water to make 1 . liter 



April, 1942] STOP AND FIXING BATH FORMULAS 371 

TABLE X 

Effect of Exhaustion with Acid Revival on the Hardening Properties of Kodak F-7 

Degree of 
Exhaustion 

(No. of Rolls Degree of Hardening 

per Gallon) Before Revival After Revival 

> 212F > 212F 

40 200 F > 212F 

80 190 F 200 F 

120 160 F 180 F 



The effect of exhaustion on the time to clear in formulas Kodak F-4 
and F-3 was similar to that obtained with Kodak F-5. The effect 
of exhaustion with formula F-7 compounded with the substitute 
hardeners Kodak F-3a and F-4a was similar to that obtained with 
Kodak F-7. 

Revival of Fixing Baths. A fixing bath is usually discarded for one 
or more of the following reasons: (a) it sludges, (b) it does not 
harden satisfactorily, (c) the time of fixation is excessive, or (d) it 
is muddy or stains the photographic material. 

In order to obtain the greatest possible use from the fixing bath, 
the acidity of the bath should be maintained relatively constant so 
that the hardening and sludging properties will not be impaired before 
the fixing power of the hypo is exhausted. 

Any of the above fixing baths containing boric acid may be revived 
with acid, using the Kodak Testing Solution A for Stop Baths in 
exactly the same manner as described above for rinse baths. Suitable 
acidic materials for revival purposes are (1) 28 per cent acetic acid, 
(2) sodium acid sulfate (Kodak SB-8R), or (3) 5 per cent sulfuric 
acid. Acids such as citric or tartaric are not suitable since they 
impair the hardening properties of the fixing bath. Also, sodium 
bisulfite is insufficiently acid for the purpose. 

Acetic acid is preferred for reviving fixing baths because a slight 
excess will not impair the properties and there is less tendency 
toward sulfurization. In all cases when reviving fixing baths with 
acid, the diluted acid should be added slowly with stirring in order to 
prevent sulfurization of the hypo. 

The effect of exhaustion with acid revival on the hardening prop- 
erties of Kodak F-7 is shown in Table X. In this test Verichrome 
film was developed in MQ-25 and fixed without rinsing. The 
acidity of the bath was then revived with 28 per cent acetic acid after 
40 rolls per gallon had been processed and the hardening determined. 



372 CRABTREE, MUEHLER, AND RUSSELL 

After an additional 40 rolls were processed the bath was again 
revived with acid and the hardening determined. Another 40 rolls 
were processed, making a total of 120 rolls per gallon, and the bath 
was again revived with acid. Samples of the bath at each stage 
were taken before and after revival and the hardening was determined . 
The results indicate that the degree of hardening decreased from a 
value greater than 212 to 200 F before revival, after processing 40 
rolls per gallon. The hardening increased to more than 212F after 
revival with acid, and then decreased to 190F after another 40 
rolls per gallon were processed. After again reviving with acid, the 
hardening increased to 200 F and then decreased to. 160 F during 
the processing of another 40 rolls per gallon. After revival with 
acid the degree of hardening was 180F. Kodak F-8 in combination 
with hardeners Kodak F-3a and F-4a could be revived in the same 
manner and the results would be similar to those obtained with 
Kodak F-7. 

Although these tests indicate that the acidity of a fixing bath can 
be satisfactorily maintained by revival, it is considered better prac- 
tice to use an intermediate acid stop bath and to maintain its acidity 
constant rather than to add acid to the fixing bath. 

Fixing baths used for paper should be tested with the Testing 
Solution B of the Kodak Testing Outfit and discarded when a posi- 
tive test is obtained. 

Revival of a used fixing bath with hypo and hardener is not gener- 
ally recommended, however, unless the photographer has available 
apparatus for adequate chemical analysis and the removal of silver. 

Choice of Acid Substitutes. The particular acid substitute which 
each individual consumer may wish to use will depend upon what 
chemicals are most readily available in his vicinity. The suggested 
order of preference is as follows : 

(A) Stop Baths 

(1) Vinegar (white) 

(2) Sodium acetate with sodium acid sulfate 
(5) Sodium bisulfite 

(4) Cifric acid 

(B) Fixing Baths 

(1) Vinegar (white) 

(2) Sodium acetate and sodium acid sulfate 

(3) Sodium bisulfite 



BOOK REVIEW 

Photography: Its Science and Practice. John R. Roebuck and Henry C. 
Staehle, D. Appleton- Century Company (New York, N. Y.), 1942. 

The present book is an outgrowth of one originally prepared by Professor Roe- 
buck for use in a course in photography in the Physics Department of the Uni- 
versity of Wisconsin about 1916. In view of the many advances that have been 
made in the art since that time, it became necessary for Professor Roebuck, 
with the aid of Dr. Henry C. Staehle of the Research Laboratories of the East- 
man Kodak Co., to rewrite the book completely. 

The result of this collaboration is a highly commendable and valuable addition 
to the photographic literature. It will prove valuable both to amateurs and 
experts, in providing an overall view of the photographic process, without going 
too deeply into technicalities. There are, in addition, thousands of persons who 
make photography their hobby and have acquired a considerable grasp of the 
technics of photography, and yet require information on the scientific funda- 
mentals of the subject. It is for such persons that the book is primarily intended. 

The whole of the modern science of photography is surveyed. A brief sketch 
of the historical development of photography leads to a study of the nature of 
the photographic emulsion and the manner in which an image is produced. At- 
tention is next given to the properties of photographic materials in relation to 
then- reaction to light, the factors determining correct exposure, and the sensi- 
tivity to color needed to preserve tonal values. The chemical processes of 
development are then treated, followed by a study of positive processes such as 
printing, enlarging, making lantern-slides, and reproduction for the graphic arts. 
The chapter on latent-image theory presents the latest views on the mechanism 
of image formation, and another chapter is devoted to lenses and the optical 
aspects of photography. Modern processes of color photography are outlined, 
and the final chapter gives a concise presentation of the subject of pictorial 
composition and aids in making good pictures. The last section of the book 
forms a brief laboratory manual consisting of a series of practical experiments. 

SYLVAN HARRIS 



373 



OFFICERS AND GOVERNORS OF THE SOCIETY 




D. E. HYNDMAN H. GRIFFIN 

Engineering V ice-President Executive V ice-President 



A. C. DOWNES 
Editorial Vice- President 




EMERY HUSE 
President 






A. S. DICKINSON 
Financial Vice-President 

374 



E. A. WlLLIFORD 

Past-President 



W. C. KUNZMANN 

Convention Vice-President 



OFFICERS AND GOVERNORS OF THE SOCIETY 



375 





P. J. LARSEN 
Secretary 



G. FRIEDL, JR. 
Treasurer 





M. C. BATSEL 
Governor 



L. L. RYDER 
Governor 





F. E. CARLSON 
Governor 



E. M. HONAN 
Governor 



J. A. MAURER 
Governor 



376 



OFFICERS AND GOVERNORS OF THE SOCIETY 





I. JACOBSEN 

Chairman, Mid-West 

Section 



A. N. GOLDSMITH 

Chairman, Atlantic 

Coast Section 



J. G. FRAYNE 

Chairman, Pacific 
Coast Section 



LOCAL SECTIONS OF THE SOCIETY 
Atlantic Coast Section 

*ALFRED N. GOLDSMITH, Chairman 

*R. O. STROCK, Past-Chairman **E. I. SPONABLE, Manager 

*SYLVAN HARRIS, Sec.-Treas. *P. C. GOLDMARK, Manager 

**C. N BATSEL, Manager *H. E. WHITE, Manager 

**M. W. PALMER, Manager *WM. H. OFFENHAUSER, JR., Manager 

Mid-West Section 

*I. JACOBSEN, Chairman 

*J. A. DUBRAY, Past-Chairman **C. H. STONE 

*L. D. STRONG, Sec.-Treas. **P. C. FOOTE 

*G. W. COLBURN, Manager *O. B. DEPUE 

*S. A. LUKES, Manager *A. SHAPIRO 

Pacific Coast Section 

*JOHN G. FRAYNE, Chairman 

*L. L. RYDER, Past-Chairman **J. K. HILLIARD, Manager 

*C. W. HANDLEY, Sec.-Treas. *B. KREUZER, Manager 

**H. W. MOYSE, Manager *S. P. SOLOW, Manager 

**J. R. WILKINSON, Manager *W. A. MUELLER, Manager 



* Term expires December 31, 1942. ** Term expires December 31. 1943. 



COMMITTEES OF THE SOCIETY 



(Correct to March 15th; additional appointments or changes may be made at any 
time during the year, as necessity or expediency may require.) 



F. E. CAHILL 
A. S. DICKINSON 



Admissions 

(East Coast) 

G. FRIEDL, JR., Chairman 
J. A. DUBRAY 
J. FRANK, JR. 



H. GRIFFIN 

D. E. HYNDMAN 



(West Coast) 
L. L. RYDER, Chairman 



J. G. FRAYNE 
K. F. MORGAN 



H. W. MOYSE 

H. W. REMERSHIED 



J. I. CRABTREE 
A. N. GOLDSMITH 



F. T. BOWDITCH 
L. E. CLARK 



Board of Editors 



A. C. DOWNES, Chairman 

A. M. GUNDELFINGER 

A. C. HARDY 
E. W. KELLOGG 

Cinema tography 

J. W. BOYLE, Chairman 

Color 

R. M. EVANS, Chairman 
J. B. ENGL 



C. R. KEITH 
C. R. SAWYER 



A. M. GUNDELFINGER 
A. C. HARDY 



J. G. FRAYNE 
J. HABER 



A. S. DICKINSON 
T. FAULKNER 
G. K. GIROUX 



Convention 

W. C. KUNZMANN, Chairman 

S. HARRIS 

H. F. HEIDEGGER 

Exchange Practice 

J. S. MACLEOD, Chairman 
G. K. HADDOW 
S. HARRIS 
L. B. ISAAC 
H. C. KAUFMAN 



O. F. NEU 
R. O. STROCK 



N. F. OAKLEY 

A. W. SCHWALBERG 
J. SlCHELMAN 



377 



378 



O. B. DEPUE 



COMMITTEES OF THE SOCIETY |j. s. M. P. E. 

Historical and Museum 

J. E. ABBOTT, Chairman 
R. GRIFFITH T. RAMSAYE 



J. I. CRABTREE 



Honorary Membership 

E. A. WILLIFORD, Chairman 
A. N. GOLDSMITH 
E. W. KELLOGG 



H. G. TASKER 



J. A. DUBRAY 



Journal Award 

R. E. FARNHAM, Chairman 

A. M. GUNDELFINGER 

O. SANDVIK 



T. E. SHEA 



J. O. BAKER 
A. C. BLANEY 
L. A. BONN 
O. B. DEPUE 
J. A. DUBRAY 
J. G. FRAYNE 



Laboratory Practice 

H. E. WHITE, Chairman 
G. H. GIBSON 
E. HUSE 
T. M. INGMAN 
C. L. LOOTENS 
A. J. MILLER 
R. F. MITCHELL 
H. W. MOYSE 



J. M. NlCKOLAUS 

N. F. OAKLEY 

W. H. OFFENHAUSER 

W. A. SCHMIDT 

J. H. SPRAY 

J. R. WILKINSON 



Membership and Subscription 

J. FRANK, JR., Chairman 

T. C. BARROWS C. W. HANDLEY 

H. BEHR S. HARRIS 

J. A. DUBRAY I. JACOBSEN 

J. G. FRAYNE G. A. JOHNSON 

E. R. GEIB W. C. KUNZMANN 

N. D. GOLDEN S. A. LUKES 



H. W. REMERSHIED 
S. P. SOLOW 
J. F. STRICKLER 
R. O. STROCK 
E. A. WILSCHKE 



J. G. BLACK 
F. E. CARLSON 
J. CLARKE 
N. B. GREEN 
F. M. HALL. 



Non-Theatrical Equipment 

J. A. MAURER, Chairman 
J. A. HAMMOND 
M. L. HOBART 
R. C. HOLSLAG 

R. KlNGSLAKE 

D. F. LYMAN 
R. F. MITCHELL 



W. H. OFFENHAUSER 
L. T. SACHTLEBEN 
A. SHAPIRO 
M. G. TOWNSLEY 
A, G, ZIMMERMAN 



April, 1942] 



F. T. BOWDITCH 
F. L. EICH 
R. E. FARNHAM 
J. L. FORREST 



COMMITTEES OF THE SOCIETY 
Papers 



379 



S. HARRIS, Chairman 
G. A. CHAMBERS, West Coast Chairman 

E. W. KELLOGG G. E. MATTHEWS 

C. R. KEITH W. H. OFFENHAUSER 

P. J. LARSEN S. P. SOLOW 

W. V. WOLFE 



J. E. ABBOTT 
J. I. CRABTREE 
A. S. DICKINSON 



Preservation of Film 

J. G. BRADLEY, Chairman 
R. M. EVANS 
J. L. FORREST 



C. L. GREGORY 
T. RAMSAYE 
V. B. SEASE 



F. R. ABBOTT 
A. H. BOLT 



Process Photography 

WILLIAM THOMAS, Chairman 
F. M. FALGE 
C. S. HANDLEY 
W. C. HOCH 



G. LAUBE 

G. H. WORRALL 



F. T. BOWDITCH 

G. L. DIMMICK 



Progress 

G. A. CHAMBERS, Chairman 

J. A. DUBRAY 

M. S. LESHING 



G. E. MATTHEWS 
D. R. WHITE 



E. W. KELLOGG 



Progress Award 

P. J. LARSEN 

E. C. RICHARDSON 



E. C. WENTE 



G. A. CHAMBERS 



Publicity 

J. HABER, Chairman 
H. A. GILBERT 



S. HARRIS 



J. O. AALBERG 
L. A. AICHOLTZ 
M. C. BATSEL 
D. G. BELL 
D. BLUMBERG 
F. E. CAHILL 
C. FLANNAGAN 



Sound 

C. R. SAWYER, Chairman 
G. FRIEDL, JR. 
E. H. HANSEN 
L. B. ISAAC 
J. P. LIVADARY 
J. A. MAURER 

R. McCULLOUGH 

B. F. MILLER 



W. C. MILLER 
K. F. MORGAN 
F. ROBERTS 
H. RUBIN 
S. SOLOW 
W. V. WOLFE 
E. C. ZRENNER 



380 



COMMITTEES OF THE SOCIETY 



[J. S. M. p. E. 



J. M. ANDREAS 
P. H. ARNOLD 
H. BARNETT 
C. N. BATSEL 
M. C. BATSEL 
F. T. BOWDITCH 
M. R. BOYER 
F. E. CARLSON 
T. H. CARPENTER 
E. K. CARVER 



Standards 

D. B. JOY, Chairman 

H. B. CUTHBERTSON 

L. W. DAVEE 
J. A. DUBRAY 
A. F. EDOUART 
J. L. FORREST 
A. N. GOLDSMITH 
H. GRIFFIN 
A. C. HARDY 
C. R. KEITH 
P. J. LARSEN 



R. LlNDERMAN 

C. L. LOOTENS 
J. A. MAURER 
G. S. MITCHELL 
W. H. OFFENHAUSER 
G. F. RACKETT 
W. B. RAYTON 
H. RUBIN 
O. SANDVIK 
H. E. WHITE 



J. W. BOYLE 
R. E. FARNHAM 



Studio Lighting 

R. LINDERMAN, Chairman 
K. FREUND 

C. W. HANDLEY 

D. B. JOY 



A. RODGERS 

K. STRAUSS 



H. BAMFORD 
R. L. CAMBPELL 
E. D. COOK 
C. E. DEAN 
J. B. ENGL 
A. N. GOLDSMITH 



Television 

P. C. GOLDMARK, Chairman 

T. T. GOLDSMITH 

H. GRIFFIN 

A. C. JENSEN 

P. J. LARSEN 

H. B. LUBCKE 

I. G. MALOFF 



J. A. MAURER 
P. MERTZ 
A. MURPHY 
O. SANDVIK 
R. E. SHELBY 
H. E. WHITE 



H. ANDERSON 
T. C. BARROWS 
H. D. BEHR 
K. BRENKERT 
F. E. CAHILL, JR. 
C. C. DASH 
A. S. DICKSINON 
J. K. ELDERKIN 
J. FRANK, JR. 



Theater Engineering 

A. N. GOLDSMITH, Chairman 

Projection Practice Sub-Committee 
C. F. HORSTMAN, Sub- Chair man 

R. R. FRENCH 

E. R. GEIB 

M. GESSIN 

A. GOODMAN 

H. GRIFFIN 

S. HARRJS 

J. J. HOPKINS 

L. B. ISAAC 

I. JACOBSEN 

I. H. LlTTENBERG 



E. R. MORIN 
J R. PRATER 

F. H. RICHARDSON 
H. RUBIN 

J. J. SEFING 
R. O. WALKER 
V. A. WELMAN 
H. E. WHITE 
A. T. WILLIAMS 



April, 1942] COMMITTEES OF THE SOCIETY 381 

Theater Design Sub- Committee 

B. SCHLANGER, Sub- Chairman 

F. W. ALEXA J. FRANK, JR. E. R. MORIN 

J. R. CLARK M. M. HARE A. L. RAVEN 

E. EBERSON S. HARRIS J. J. SEFING 

C. F. HORSTMAN 

Screen Brightness Sub- Committee 
F. E. CARLSON, Sub- Chair man 

H. BARNETT W. F. LITTLE C. TUTTLE 

E. R. GEIB W. B. RAYTON H. E. WHITE 

S. HARRIS A. T. WILLIAMS 

SMPE REPRESENTATIVES TO OTHER ORGANIZATIONS 

American Documentation Institute J. E. ABBOTT 

Sectional Committee on Motion Pictures, ASA E, K. CARVER 

A. N. GOLDSMITH 
H. G. TASKER 

Sectional Committee on Photography, ASA J. I. CRABTREE 

Inter-Society Color Council R. M. EVANS 

J. A. BALL 

F. T. BOWDITCH 

G. F. RACKETT 
Sectional Committee on Standardization of Letter Symbols L. A. JONES 

and Abbreviations for Science and Engineering, ASA 
National Association of Broadcasters, Committee on Re- R. M. MORRIS 
cording and Reproducing Standards 

American Standards Association 
Sectional Committee on Motion Pictures (Z22) 

ALFRED N. GOLDSMITH, Chairman 
SYLVAN HARRIS, Chairman 



J. O. AALBERG 


F. EDOUART 


G. S. MITCHELL 


P. H. ARNOLD 


E. W. ELY 


O. F. NEU 


M. C. BATSEL 


R. E. FARNHAM 


N. F. OAKLEY 


C. R. BROWN 


H. GRIFFIN 


*(M. R. BOYER) 


B. H. CARROLL 


R. G. HOLSLAG 


D. PALFREYMAN 


E. K. CARVER 


L. A. JONES 


A. R. SMALL 


KENNETH CLARK 


D. B. JOY 


*(G. W. BOOTH) 


W. CLARK 


*(E. A. WILLIFORD) 


J. L. SPENCE 


A. S. DICKINSON 


C. R. KEITH 


H. G. TASKER 


J. A. DUBRAY 


G. A. MITCHELL 


G. H. WORRALL 




*(J. RUTTENBERG) 





* Alternate. 



CONSTITUTION AND BY-LAWS 

OF THE 
SOCIETY OF MOTION PICTURE ENGINEERS* 

CONSTITUTION 
Article 1 

Name 

The name of this association shall be SOCIETY OF MOTION PICTURE 
ENGINEERS. 

Article II 

Object 

Its objects shall be: Advancement in the theory and practice of motion pic- 
ture engineering and the allied arts and sciences, the standardization of the equip- 
ment, mechanisms, and practices employed therein, the maintenance of a high 
professional standing among its members, and the dissemination of scientific 
knowledge by publication. 

Article III 

Eligibility 

Any person of good character may be a member in any grade for which he is 
eligible. 

Article IV 

Officers 

The officers of the Society shall be a President, a Past-President, an Executive 
Vice-President, an Engineering Vice-President, an Editorial Vice-President, a 
Financial Vice-President, a Convention Vice-President, a Secretary, and a 
Treasurer. 

The term of office of the President, the Past-President, the Executive Vice- 
President, the Engineering Vice-President, the Editorial Vice-President, the 
Financial Vice-President, and the Convention Vice-President shall be two years, 
and the Secretary and the Treasurer one year. Of the Engineering, Editorial. 
Financial, and Convention Vice-Presidents, two shall be elected alternately each 
year, or until their successors are chosen. The President shall not be immediately 
eligible to succeed himself in office. 

Article V 

Board of Governors 

The Board of Governors shall consist of the President, the Past-President, the 
five Vice-Presidents, the Secretary, the Treasurer, the Section Chairmen, and 

* Corrected to March 15. 1941. 
382 



CONSTITUTION AND BY-LAWS 383 

five elected Governors. Two, and three, of the Governors shall be elected al- 
ternately each year to serve for two years. 

Article VI 

Meetings 

There shall be an annual meeting, and such other meetings as stated in the 
By-Laws. 

Article VII 

Amendments 

This Constitution may be amended as follows : Amendments shall be approved 
by the Board of Governors, and shall be submitted for discussion at any regular 
members' meeting. The proposed amendment and complete discussion then 
shall be submitted to the entire Active, Fellow, and Honorary membership, 
together with letter ballot as soon as possible after the meeting. Two-thirds of 
the vote cast within sixty days after mailing shall be required to carry the amend- 
ment. 

BY-LAWS 
By-Law I 

Membership 

Sec. 1. The membership of the Society shall consist of Honorary members, 
Fellows, Active members, Associate members, and Sustaining members. 

An Honorary member is one who has performed eminent services in the ad- 
vancement of motion picture engineering or in the allied arts. An Honorary 
member shall be entitled to vote and to hold any office in the Society. 

A Fellow is one who shall not be less than thirty years of age and who shall 
comply with the requirements of either (a) or (b) for Active members and, in 
addition, shall by his proficiency and contributions have attained to an out- 
standing rank among engineers or executives of the motion picture industry. 
A Fellow shall be entitled to vote and to hold any office in the Society. 

An Active member is one who shall be not less than 25 years of age, and shall 
be: 

. (a) A motion picture engineer by profession. He shall have been engaged in 
the practice of his profession for a period of at least three years, and shall have 
taken responsibility for the design, installation, or operation of systems or ap- 
paratus pertaining to the motion picture industry. 

(b) A person regularly employed in motion picture or closely allied work, 
who by his inventions or proficiency in motion picture science or as an executive 
of a motion picture enterprise of large scope, has attained to a recognized stand- 
ing in the motion picture industry. In case of such an executive, the applicant 
must be qualified to take full charge of the broader features of motion picture 
engineering involved in the work under his direction. 

(c) An Active member is privileged to vote and to hold any office in the So- 
ciety. 

An Associate member is one who shall be not less than 18 years of age, and shall 
be a person who is interested in or connected with the study of motion picture 



384 CONSTITUTION AND BY-LAWS [J. S. M. P. E. 

technical problems or the application of them. An Associate member is not privi- 
leged to vote, to hold office or to act as chairman of any committee, although he 
may serve upon any committee to which he may be appointed; and, when so' 
appointed, shall be entitled to the full voting privileges of a committee member. 

(d) A student member is any person registered as a student, graduate or under- 
graduate, in a college, university, or educational institution, pursuing a course of 
studies in science or engineering that evidences interest in motion picture tech- 
nology. Membership in this grade shall not extend more than one year beyond 
the termination of the student status described above. A student member shall 
have the same privileges as Associate members of the Society. 

A Sustaining member is an individual, a firm, or corporation contributing 
substantially to the financial support of the Society. 

Sec. 2. All applications for membership or transfer, except for honorary or 
fellow membership, shall be made on blank forms provided for the purpose, and 
shall give a complete record of the applicant's education and experience. Honor- 
ary and Fellow membership may not be applied for. 

Sec. 3. (a) An Honorary membership may be granted upon recommendation 
of the Board of Governors when confirmed by a four-fifths majority vote of the 
Honorary members, Fellows, and Active members present at any regular meeting 
of the Society. An Honorary member shall be exempt from all dues. 

(6) Fellow membership may be granted upon recommendation of the Fellow 
Membership Award Committee, when confirmed by a three-fourths majority vote 
of the Board of Governors. 

(c) Applicants for Active Membership shall give as reference at least three 
members of Active or of higher grade in good standing. Applicants shall be elected 
to membership by the unanimous approval of the entire membership of the ap- 
propriate Admissions Committee. In the event of a single dissenting vote or 
failure of any member of the Admissions Committee to vote, the application shall 
be referred to the Board of Governors, in which case approval of at least three- 
fourths of the Board of Governors shall be required. 

(d) Applicants for Associate membership shall give as reference at least one 
member of higher grade in good standing. Applicants shall be elected to member- 
ship by approval of a majority of the appropriate Admissions Committee. 

(e) Applicants for student membership shall give as reference the head of the 
Department of the Institution he is attending; this faculty member not neces- 
sarily being a member of the Society. 



By-Law II 

Officers 

Sec. 1. An officer or governor shall be an Honorary, a Fellow, or Active mem- 
ber. 

Sec. 2. Vacancies in the Board of Governors shall be filled by the Board of 
Governors until the annual meeting of the Society, 



April, 1942] CONSTITUTION AND BY-LAWS 385 

By-Law III 

Board of Governors 

Sec. 1. The Board of Governors shall transact the business of the Society 
between members' meetings, and shall meet at the call of the president. 

Sec. 2. A majority of the Board of Governors shall constitute a quorum at 
regular meetings. 

Sec. 3. When voting by letter ballot, a majority affirmative vote of the total 
membership of the Board of Governors shall carry approval, except as otherwise 
provided. 

Sec. 4. The Board of Governors, when making nominations to office, and to 
the Board, shall endeavor to nominate persons, who in the aggregate are repre- 
sentative of the various branches or organizations of the motion picture industry, 
to the end that there shall be no substantial predominance upon the Board, as the 
result of its own action, of representatives of any one or more branches or organi- 
zations of the industry. 

By- Law IV 
Committees 

Sec. 1. All committees, except as otherwise specified, shall be appointed by the 
President. 

Sec. 2. All committees shall be appointed to act for the term served by the 
officer who shalkappoint the committees, unless their appointment is sooner ter- 
minated by the appointing officer. 

Sec. 3. Chairman of the committees shall not be eligible to serve in such ca- 
pacity for more than two consecutive terms. 

Sec. 4. Standing committees of the Society shall be as follows to be appointed 
as designated: 

(a) Appointed by the President and confirmed by the Board of Governors 
Progress Award Committee 
Journal Award Committee 
Honorary Membership Committee 
Fellow Membership Award Committee 
Admissions Committees 

(Atlantic and Mid- West Sections) 
(Pacific Coast Section) 
European Advisory Committee 
(6) Appointed by the Engineering Vice-president 
Sound Committee 
Standards Committee 
Studio Lighting Committee 
Color Committee 
Theater Engineering Committee 
Exchange Practice Committee 
Non-Theatrical Equipment Committee 
Television Committee 
Laboratory Practice Committee 
Committee on Cinematography 
Process Photography Committee 
Committee on Preservation of Film 



386 CONSTITUTION AND BY-LAWS [J. S. M. P. E. 

(c) Appointed by Editorial Vice- President 

Board of Editors 
Papers Committee 
Progress Committee 
Historical Committee 
Museum Committee 

(d) Appointed by Convention Vice- President 

Publicity Committee 
Convention Arrangements Committee 
Apparatus Exhibit Committee 
() Appointed by Financial Vice- President 

Membership and Subscription Committee 

Sec. 5. Two Admissions Committees, one for the Atlantic and Mid- West 
Sections, and one for the Pacific Coast Section, shall be appointed. The former 
committee shall consist of a chairman and six Fellow or Active members of the 
Society of which four shall be members of the Board of Governors. The latter 
committee shall consist of a Chairman and four Fellow or Active members of the 
Society including all officers or members of the Board of Governors of the Society 
residing in the Pacific Coast Section. 

By-Law V 

Meetings 

Sec. 1. The location of each meeting of the Society shall be determined by the 
Board of Governors. 

Sec. 2. Only Honorary members, Fellows, and Active members shall be en- 
titled to vote. 

Sec. 3. A quorum of the Society shall consist in number of one-tenth of the 
total number of Honorary members, Fellows, and Active members as listed in 
the Society's records at the close of the last fiscal year. 

Sec. 4. The fall convention shall be the annual meeting. 

Sec. 5. Special meetings may be called by the president and upon the request 
of any three members of the Board of Governors not including the president. 

Sec. 6. All members of the Society in any grade shall have the privilege of 
discussing technical material presented before the Society or its Sections. 

By-Law VI 

Duties of Officers 

Sec. 1. The president shall preside at all business meetings of the Society and 
shall perform the duties pertaining to that office. As such he shall be the chief 
executive of the Society, to whom all other officers shall report. 

Sec. 2. In the absence of the president, the officer next in order as listed in 
Article 4 of the Constitution shall preside at meetings and perform the duties of 
the president. 

Sec. 3. The five vice-presidents shall perform the duties separately enumerated 
below for each office, or as denned by the president : 

(a) The executive vice-president shall represent the president in such geo- 
graphical areas of the United States as shall be determined by the Board of Gover- 



April, 1942] CONSTITUTION AND BY-LAWS 387 

nors, and shall be responsible for the supervision of the general affairs of the 
Society in such areas, as directed by the president of the Society. 

(b) The engineering vice-president shall appoint all technical committees. 
He shall be responsible for the general initiation, supervision, and coordination of 
the work in and among these committees. He may act as chairman of any com- 
mittee or otherwise be a member ex-officio. 

(c) The editorial vice-president shall be responsible for the publication of the 
Society's JOURNAL and all other technical publications. He shall pass upon the 
suitability of the material for publication, and shall cause material suitable for 
publication to be solicited as may be needed. He shall appoint a papers com- 
mittee and an editorial committee. He may act as chairman of any committee or 
otherwise be a member ex-officio. 

( d) The financial vice-president shall be responsible for the financial operations 
of the Society, and shall conduct them in accordance with budgets approved by 
the Board of Governors. He shall study the costs of operation and the income 
possibilities to the end that the greatest service may be rendered to the members 
of the Society within the available funds. He shall submit proposed budgets to 
the Board. He shall appoint at his discretion a ways and means committee, a 
membership committee, a commercial advertising committee, and such other 
committees within the scope of his work as may be needed. He may act as chair- 
man of any of these committees or otherwise be a member ex-officio. 

(e) The convention vice-president shall be responsible for the national con- 
ventions of the society. He shall appoint a convention arrangements committee, 
an apparatus exhibit committee, and a publicity committee. He may act as 
chairman of any committee, or otherwise be a member ex-officio. 

Sec. 4. The secretary shall keep a record of all meetings; he shall conduct the 
correspondence relating to his office, and shall have the care and custody of 
records, and the seal of the Society. 

Sec. 5. The treasurer shall have charge of the funds of the Society and disburse 
them as and when authorized by the financial vice-president. He shall make 
an annual report, duly audited, to the Society, and a report at such other times 
as may be requested. He shall be bonded in an amount to be determined by the 
Board of Governors and his bond filed with the Secretary. 

Sec. 6. Each officer of the Society, upon the expiration of his term of office, 
shall transmit to his successor a memorandum outlining the duties and policies 
of his office. 

By-Law VII 

Elections 

Sec. 1. (a) All officers and five governors shall be elected to their respective 
offices by a majority of ballots cast by the Active, Fellow, and Honorary members 
in the following manner: 

Not less than three months prior to the annual fall convention, the Board of 
Governors shall nominate for each vacancy several suitable candidates. Nomi- 
nations shall first be presented by a Nominating Committee appointed by the 
President, consisting of nine members, including a chairman. The committee 
shall be made up of two Past-Presidents, three members of the Board of Governors 
not up for election, and four other Active, Fellow, or Honorary members, not 



388 CONSTITUTION AND BY-LAWS [J. S. M. p. E. 

currently officers or Governors of the Society. Nominations shall be made by 
three-quarters affirmative vote of the total Nominating Committee. Such nomi- 
nations shall be final unless any nominee is rejected by a three-quarters vote of 
the Board of Governors present and voting. 

The secretary shall then notify these candidates of their nomination. From 
the list of acceptances, not more than two names for each vacancy shall be se- 
lected by the Board of Governors and placed on a letter ballot. A blank space 
shall be provided on this letter ballot under each office, in which space the names 
of any Active, Fellow, or Honorary members other than those suggested by the 
Board of Governors may be voted for. The balloting shall then take place. 

The ballot shall be enclosed hi a blank envelope which is enclosed in an outer 
envelope bearing the secretary's address and a space for the member's name and 
address. One of these shall be mailed to each Active, Fellow, and Honorary 
member of the Society, not less than forty days in advance of the annual fall 
convention. 

The voter shall then indicate on the ballot one choice for each office, seal the 
ballot in the blank envelope, place this in the envelope addressed to the secretary, 
sign his name and address on the letter, and mail it in accordance with the in- 
structions printed on the ballot. No marks of any kind except those above pre- 
scribed shall be placed upon the ballots or envelopes. 

The sealed envelope shall be delivered by the secretary to a committee of 
tellers appointed by the president at the annual fall convention. This com- 
mittee shall then examine the return envelopes, open and count the ballots, and 
announce the results of the election. 

The newly elected officers and governors of the general Society shall take office 
on the January 1st following their election. 

(6) The first group of vice-presidents, viz., the executive vice-president, 
engineering vice-president, editorial vice-president, financial vice-president, con- 
vention vice-president, and a fifth governor, shall be nominated by the Board of 
Governors at its first meeting after the ratification of the corresponding provisions 
of the Constitution; and the membership shall vote on the candidates in ac- 
cordance with the procedure prescribed in these By-Laws for regular elections of 
officers so far as these may be applicable. 

By-Law VIII 

Dues and Indebtedness 

Sec. 1. The annual dues shall be fifteen dollars ($15) for Fellows and Active 
members, seven dollars and fifty cents ($7.50) for Associate members, and three 
dollars ($3.00) for Student members, payable on or before January 1st of each 
year. Current or first year's dues for new members, dating from the notification 
of acceptance in the Society, shall be prorated on a monthly basis. Five dollars 
of these dues shall apply for annual subscription to the JOURNAL. No admission 
fee will be required for any grade of membership. 

Sec. 2. (a) Transfer of membership may be made effective at any time by pay- 
ment of the pro rata dues for the current year. 

(6) No credit shall be given for annual dues in a membership transfer from a 
higher to a lower grade, and such transfers shall take place on January 1st of each 
year. 



April, 1942] CONSTITUTION AND BY-LAWS 389 

(c) The Board of Governors upon their own initiative and without a transfer 
application may elect, by the approval of at least three-fourths of the Board, any 
Associate or Active member for transfer to any higher grade of membership. 

Sec. 3. Annual dues shall be paid in advance. All Honorary members, Fel- 
lows, and Active members in good standing, as denned in Section 5, may vote or 
otherwise participate in the meetings. 

Sec. 4. Members shall be considered delinquent whose annual dues for the 
year remain unpaid on February 1st. The first notice of delinquency shall be 
mailed February 1st. The second notice of delinquency shall be mailed, if neces- 
sary, on March 1st, and shall include a statement that the member's name will be 
removed from the mailing list for the JOURNAL and other publications of the 
Society before the mailing of the April issue of the JOURNAL. Members who are 
in arrears of dues on June 1st, after two notices of such delinquency have been 
mailed to their last address of record, shall be notified their names have been re- 
moved from the mailing list and shall be warned unless remittance is received on or 
before August 1st, their names shall be submitted to the Board of Governors for 
action at the next meeting. Back issues of the JOURNAL shall be sent, if available, 
to members whose dues have been paid prior to August 1st. 

Sec. 5. (a) Members whose dues remain unpaid on October 1st may be 
dropped from the rolls of the Society by majority vote and action of the Board 
or the Board may take such action as it sees fit. 

(6) Anyone who has been dropped from the rolls of the Society for non-pay- 
ment of dues shall, in the event of his application for reinstatement, be considered 
as a new member. 

(c) Any member may be suspended or expelled for cause by a majority vote of 
the entire Board of Governors; provided he shall be given notice and a copy in 
writing of the charges preferred against him, and shall be afforded opportunity 
to be heard ten days prior to such action. 

Sec. 6. The provisions of Sections 1 to 4, inclusive, of this By-Law VIII given 
above may be modified or rescinded by action of the Board of Governors. 



By -Law IX 

Emblem 

Sec. 1. The emblem of the Society shall be a facsimile of a four-hole film -reel 
with the letter S in the upper center opening, and the letters M, P, and E, in the 
three lower openings, respectively. The Society's emblem may be worn by 
members only. 

By-Law X 

Publications 

Sec. 1. Papers read at meetings or submitted at other times, and all material 
of general interest shall be submitted to the editorial board, and those deemed 
worthy of permanent record shall be printed in the JOURNAL. A copy of each issue 
shall be mailed to each member in good standing to his last address of record. 
Extra copies of the JOURNAL shall be printed for general distribution and may be 



390 CONSTITUTION AND BY-LAWS [J. S. M. P. E. 

obtained from the General Office on payment of a fee fixed by the Board of 
Governors. 

By-Law XI 

Local Sections 

Sec. 1 . Sections of the Society may be authorized in any state or locality where 
the Active, Fellow, and Honorary membership exceeds 20. The geographic 
boundaries of each Section shall be determined by the Board of Governors. 

Upon written petition, signed by 20 or more Active members, Fellows and Hon- 
orary members, for the authorization of a Section of the Society, the Board of 
Governors may grant such authorization. 

Membership 

Sec. 2. All members of the Society of Motion Picture Engineers in good stand- 
ing residing in that portion of any country set apart by the Board of Governors 
tributary to any local Section shall be eligible for membership in that Section, and 
when so enrolled they shall be entitled to all privileges that such local Section may, 
under the General Society's Constitution and By-Laws, provide. 

Any member of the Society in good standing shall be eligible for non-resident 
affiliated membership of any Section under conditions and obligations prescribed 
for the Section. An affiliated member shall receive all notices and publications 
of the Section but he shall not be entitled to vote at Sectional Meetings. 

Sec. 3. Should the enrolled Active, Fellow, and Honorary membership of a 
Section fall below 20, or should the technical quality of the presented papers fall 
below an acceptable level, or the average attendance at meetings not warrant the 
expense of maintaining the organization, the Board of Governors may cancel its 
authorization. 

Officers 

Sec. 4. The officers of each section shall be a chairman and a secretary- 
treasurer. The Section chairmen shall automatically become members of the 
Board of Governors of the General Society, and continue in such positions for the 
duration of their terms as chairmen of the local sections. Each Section officer 
shall hold office for one year, or until his successor is chosen. 

Board of Managers 

Sec. 5. The Board of Managers shall consist of the Section chairman, the 
Section past-chairman, the Section secretary-treasurer, and six Active, Fellow, or 
Honorary members. Each manager of a Section shall hold office for two years, 
or until his successor is chosen. 

Elections 

Sec. 6. The officers and managers of a Section shall be Active, Fellow, or 
Honorary members of the General Society. 

Not less than three months prior to the annual Fall Convention of the Society, 



April, 1942] CONSTITUTION AND BY-LAWS 391 

nominations shall be presented to the Board of Managers of the Section by a 
Nominating Committee appointed by the chairman of the Section, consisting of 
seven members, including a chairman. The Committee shall be composed of the 
present chairman, the past-chairman, two other members of the Board of Man- 
agers not up for election, and three other Active, Fellow, or Honorary members of 
the Section not currently officers or managers of the Section. Nominations shall 
be made by a three-quarters affirmative vote of the total Nominating Committee. 
Such nominations shall be final, unless any nominee is rejected by a three-quarters 
vote of the Board of Managers, and in the event of such rejection the Board of 
Managers will make its own nomination. 

The Chairman of the Section shall then notify these candidates of their nomi- 
nation. From the list of acceptances, not more than two names for each vacancy 
shall be selected by the Board of Managers and placed on a letter ballot. A blank 
space shall be provided on this letter ballot under each office, in which space the 
names of any Active, Fellow, or Honorary members other than those suggested 
by the Board of Managers may be voted for. The balloting shall then take place. 

The ballot shall be enclosed in a blank envelope which is enclosed in an outer 
envelope bearing the local Secretary-Treasurer's address and a space for the 
member's name and address. One of these shall be mailed to each Active, 
Fellow, and Honorary member of the Society, residing in the geographical area 
covered by the Section, not less than forty days in advance of the annual Fall 
Convention. 

The voter shall then indicate on the ballot one choice for each office, seal the 
ballot in the blank envelope, place this in the envelope addressed to the Secretary- 
Treasurer, sign his name and address on the letter, and mail it in accordance with 
the instructions printed on the ballot. No marks of any kind except those above 
prescribed shall be placed upon the ballots or envelopes. 

The sealed envelopes shall be delivered by the Secretary-Treasurer to his 
Board of Managers at a duly called meeting. The Board of Managers shall then 
examine the return envelopes, open and count the ballots, and announce the 
results of the election. 

The newly elected officers and managers shall take office on the January 1st 
following their election. 

Business 
Sec. 7. The business of a Section shall be conducted by the Board of Managers. 

Expenses 

Sec. 8. (a) As early as possible in the fiscal year, the secretary of each Section 
shall submit to the Board of Governors of the Society a budget of expenses for the 
year. 

(6) The treasurer of the General Society may deposit with each Section secre- 
tary-treasurer a sum of money, the amount to be fixed by the Board of Governors, 
for current expenses. 

(c) The secretary-treasurer of each Section shall send to the treasurer of the 
General Society, quarterly or on demand, an itemized account of all expenditures 
incurred during the preceding interval. 



392 CONSTITUTION AND BY-LAWS 

(d) Expenses other than those enumerated in the budget, as approved by the 
Board of Governors of the General Society, shall not be payable from the general 
funds of the Society without express permission from the Board of Governors. 

(e) A Section Board of Managers shall defray all expenses of the Section not 
provided for by the Board of Governors, from funds raised locally by donation, 
or fixed annual dues, or by both. 

(/) The secretary of the Society shall, unless otherwise arranged, supply to 
each Section all stationery and printing necessary for the conduct of its business. 

Meetings 

Sec. 9. The regular meetings of a Section shall be held in such places and at such 
hours as the Board of Managers may designate. 

The secretary-treasurer of each Section shall forward to the secretary of the 
General Society, not later than five days after a meeting of a Section, a statement 
of the attendance and of the business transacted. 

Papers 

Sec. 10. Papers shall be approved by the Section's papers committee previ- 
ously to their being presented before a Section. Manuscripts of papers presented 
before a Section, together with a report of the discussions and the proceedings of 
the Section meetings, shall be forwarded promptly by the Section secretary- 
treasurer to the secretary of the General Society. Such material may, at the dis- 
cretion of the Board of Editors of the General Society, be printed in the Society's 
publications. 

Constitution and By-Laws 

Sec. 11. Sections shall abide by the Constitution and By-Laws of the Society 
and conform to the regulations of the Board of Governors. The conduct of Sec- 
tions shall always be in conformity with the general policy of the Society as fixed 
by the Board of Governors. 

By-Law XII 

Amendments 

Sec. 1. These By-Laws may be amended at any regular meeting of the Society 
by the affirmative vote of two-thirds of the members present at a meeting who are 
eligible to vote thereon, a quorum being present, either on the recommendation of 
the Board of Governors or by a recommendation to the Board of Governors signed 
by any ten members of active or higher grade, provided that the proposed amend- 
ment or amendments shall have been published in the JOURNAL of the Society, 
in the issue next preceding the date of the stated business meeting of the Society 
at which the amendment or amendments are to be acted upon. 

Sec. 2. In the event that no quorum of the voting members is present at the 
time of the meeting referred to in Section 1, the amendment or amendments shall 
be referred for action to the Board of Governors. The proposed amendment or 
amendments then become a part of the By-Laws upon receiving the affirmative 
vote of three-quarters of the Board of Governors. 



FIFTY-FIRST SEMI-ANNUAL CONVENTION 
OF THE 

SOCIETY OF MOTION PICTURE ENGINEERS 



HOLLYWOOD-ROOSEVELT HOTEL 

HOLLYWOOD, CALIF. 
MAY 4th-8th, INCLUSIVE 



OFFICERS AND COMMITTEES IN CHARGE 

EMERY HUSE, President 

E. A. WILLIFORD, Past-President 

H. GRIFFIN, Executive V ice-President 

W. C. KUNZMANN, Convention Vice-President 

A. C. DOWNES, Editorial Vice-President 

J. G. FRAYNE, Chairman, Pacific Coast Section 

C. W. HANDLEY, Chairman, Local Arrangements Committee 

S. HARRIS, Chairman, Papers Committee 

Pacific Coast Papers Committee 



G. A. CHAMBERS 
C. R. DAILY 



R. R. SCOVILLE, Chairman 
F. L. EICH 
W. W. LINDSAY, JR. 



S. P. SOLOW 
W. V. WOLFE 



Reception and Local Arrangements 



J. O. AALBERG 
B. B. BROWN 
G. A. CHAMBERS 
W. E. GARITY 

A. M. GUNDELFINGER 

E. H. HANSEN 

J. K. HlLLIARD 

E. M. HONAN 



C. W. HANDLEY, Chairman 

B. KREUZER 

R. G. LlNDERMAN 

C. L. LOOTENS 

R. H. McCULLOUGH 

W. C. MILLER 
G. S. MITCHELL 
K. F. MORGAN 
H. MOYSE 



W. A. MUELLER 
G. F. RACKETT 
H. W. REMERSHIED 
ALSTON RODGERS 
L. L. RYDER 
S. P. SOLOW 
H. G. TASKER 
J. R. WILKINSON 



Registration and Information 

W. C. KUNZMANN, Chairman 
F. ALBIN J. FRANK, JR. 

L. W. CHASE J. G. FRAYNE 

C. W. HANDLEY 



S. HARRIS 
F. L. HOPPER 



393 



394 



L. A. AlCHOLTZ 

J. W. BOYLE 
J. L. COURCIER 



1942 SPRING CONVENTION 

Publicity 

JULIUS HABER, Chairman 
G. R. GIROUX, West Coast Chairman 
S. HARRIS 
S. E. HAWKINS 



[J. S. M. P. E. 



G. S. MITCHELL 
E. C. RICHARDSON 
R. R. SCOVILLE 



Luncheon and Banquet Committee 

L. L. RYDER, Chairman 
J O. AALBERG EMERY HUSE 

J. G. FRAYNE H. T. KALMUS 

C. W. HANDLEY M. S. LESHING 

E. M. HONAN N. LEVINSON 



R. H. McCULLOUGH 

W. C. MILLER 

P. MOLE 

H. G. TASKER 



A. C. BLANEY 
D. J. BLOOMBERG 
L. F. BROWN 
J. P. CORCORAN 



J. O. AALBERG 

J. DURST 

G. M. FARLY 

B. FREERICKS 

W. E. GEBHART, JR. 



Hotel and Transportation 

G. A. CHAMBERS, Chairman 
C. R. DAILY 
C. DUNNING 
W. C. HARCUS 
G. T. LORANCE 

Convention Projection 

C. L. RUSSELL, Chairman 
L. D. GRIGNON 
J. K. HILLIARD 
A. E. JACKSON 
W. W. LINDSAY, JR. 

R. H. McCULLOUGH 



H. R. LUBCKE 

F. O'GRADY 

J. W. STAFFORD 
W. L. THAYER 



S. M. PARISEAU 
H. W. REMERSHIED 
C. R. SAWYER 
G. E. SAWYER 
H. A. STARKE 



Officers and Members of Los Angeles Projectionists Local No. 160 



Ladies' Reception Committee 

MRS. EMERY HUSE and MRS. J. G. FRAYNE, Hostesses 



Assisted by 



MRS. G. A CHAMBERS 
MRS. F. L. EICH 

MRS. A. M. GUNDELFINGER 

MRS. C. W. HANDLEY 
MRS. J. K. HILLIARD 
MRS. E. M. HONAN 
MRS. B. KREUZER 
MRS. N. LBVINSON 
MRS. R. H. MCCULLOUGH 
MRS. G. S. MITCHELL 



MRS. 
MRS. 
MRS. 
MRS. 
MRS. 
MRS. 
MRS. 
MRS. 
MRS. 
MRS. 



P. MOLE 

K. F. MORGAN 

W. A. MUELLER 

G. F. RACKETT 

H. W. REMERSHIED 

E. C. RICHARDSON 

L. L. RYDER 

R. R. SCOVILLE 

S. P. SOLOW 

J. R. WILKINSON 



MRS. W. V. WOLFE 



April, 1942] 1942 SPRING CONVENTION 395 

Color Print Exhibit Committee 

O. O. CECCARINI, Chairman 

L. E. CLARK C. DUNNING L. D. GRIGNON 

T. B. CUNNINGHAM R. M. EVANS A. M. GUNDELFINGER 

TENTATIVE PROGRAM 

MONDAY, MAY 4, 1942 

9:00 a.m. Hotel Lobby; Registration 
12:30 p.m. Terrace Room; Informal Get-Together Luncheon 

Addresses by prominent Hollywood members of the motion picture 
industry; names to be announced later 
2:00 Blossom Room; General Session 

8:00 Blossom Room; Technical Session 

TUESDAY, MAY 5, 1942 

9:30 a.m. Hotel Lobby; Registration 

This morning will be left open for a possible trip or other activity to 
be announced later 

2 : 00 p.m. Blossom Room; Technical Session 
8:00 Blossom Room; Technical Session 

WEDNESDAY, MAY 6, 1942 

9:30 a.m. Hotel Lobby; Registration 
10:00 Blossom Room; Technical Session 

2:00 p.m. The afternoon will be left open for a possible trip, to be announced 

later, or for recreation 
8:30 Blossom Room; Fifty-First Semi- Annual Banquet and Dance; details 

to be announced later 

THURSDAY, MAY 7, 1942 

10:30 a.m. Hotel Lobby; Registration 

Open morning 

2:00 p.m. Blossom Room; Technical Session 
8:00 Blossom Room; Technical Session 

FRIDAY, MAY 8, 1942 

10:00 a.m. Blossom Room; Technical Session 
2:00 p.m. Blossom Room; Technical Session 
8:00 Blossom Room; Technical Session 

Adjournment of the Convention 



396 1942 SPRING CONVENTION 

HEADQUARTERS 

The Convention headquarters will be at the Hollywood-Roosevelt Hotel. 
Excellent accommodations have been assured by the hotel management at the 
following per diem rates: 

One person, room and bath $3 . 60 

Two persons, double bed and bath 5.00 

Two persons, twin beds and bath 6 . 00 

Parlor suite and bath, one person 8.00-14.00 

Parlor suite and bath, two persons 12.00-16.00 

Room reservation cards will be mailed to the membership early in April and 
should be returned to the hotel immediately to be assured of satisfactory ac- 
commodations. 

Indoor and outdoor parking facilities adjacent to the hotel will be available for 
those who motor to the Convention. 

Golfing privileges may be arranged by request of the hotel management or at 
the registration headquarters. 

Registration headquarters will be in the hotel lobby. All members and guests 
attending the Convention will be expected to register and receive their Conven- 
tion badges. The registration fees are used to defray the expenses of the Con- 
vention, and cooperation in this respect is requested. Identification cards will 
be supplied, which will serve as admittance to all scheduled or special sessions, 
studio visits, and trips, and several de luxe motion picture theaters on Hollywood 
Boulevard in the vicinity of the hotel. 

Members planning to attend the Convention should consult their local railroad 
passenger agents regarding train schedules, rates, and stop-over privileges en 
route. For a stop-over at San Francisco the Convention Committee recommends 
the Mark Hopkins Hotel, on "Nob Hill." Accommodations may be arranged 
with Mr. George D. Smith, manager of this hotel. 

An interesting color-print exhibit will be an adjunct to the Convention and will 
be open to the public and delegates during the five days of the Convention. 

The Convention hostesses promise an interesting program of entertainment for 
the visiting ladies. A reception parlor will be provided as their headquarters at 
the hotel. 

Note: The Pacific Coast Section officers and managers gave serious considera- 
tion to the question of holding the 1942 Spring Convention at Hollywood, and 
have decided to proceed with arrangements for the meeting. The motion picture 
industry plays an essential part from the exhibiting and engineering viewpoint in 
upholding the morale of the general and theater-going public in the present crisis, 
and accordingly the Convention and Local Arrangements Committees are por- 
ceeding with their plans. However, if later deemed advisable in the National 
interest, the Convention will be subject to cancellation thirty days prior to the 
announced Convention dates. 

W. C. KUNZMANN, 

Convention Vice-President 



SOCIETY ANNOUNCEMENTS 
FIFTY-FIRST SEMI-ANNUAL CONVENTION 



HOLLYWOOD-ROOSEVELT HOTEL, HOLLYWOOD, CALIF. 
MAY 4th-8th, INCLUSIVE 



Plans for the approaching Convention at Hollywood are going forward rapidly 
and a very fine program of papers will be available. Nine sessions will be devoted 
to technical presentations, four of them in the evenings to make it possible for 
those to be present who are engaged at the studios during the daytime. 

Details concerning the meetings will be found in another section of this issue 
of the JOURNAL. The Tentative Program listing the papers and presentations 
will be mailed to the membership of the Society about the middle of this month, 
together with hotel reservation cards. 

Members who contemplate attending the Convention should not delay in 
returning their reservation cards, so as to be assured of satisfactory accommoda- 
tions in the hotel. 

ATLANTIC COAST SECTION 

On February 19th, at a meeting held at the Hotel Pennsylvania, New York, 
Messrs. R. Blackinton Fuller and L. S. Rhodes of the Marsh Cinesound, Inc., 
presented a paper on "Procedural and Dimensional Practices for Production of 
16-Mm Motion Pictures for Television Projection." A very interesting and 
lively discussion followed the presentation. 

On March 19th at the New Yorker Hotel, Messrs. A. T. Williams and Wm. 
A. R. Reedy of the Weston Electric Instrument Corp., Newark, N. J., presented 
a talk on the "Characteristics, Design, and Use of Photoelectric Exposure 
Meters." The discussion was divided into two sections, the first covering 
the characteristics and design of present-day exposure meters, methods of cali- 
brating meters, and the use of reflected rather than incident light as a criterion 
for exposure. 

The second section was devoted to the practical uses of the photoelectric ex- 
posure meter, illustrated by slides in both black-and-white and color. 

On April 16th, Mr. R. M. Evans of the Eastman Kodak Co., Rochester, N. Y., 
will present a paper on the present-day status of still-color photography, includ- 
ing Kodacolor, Minicolor, Kotavachrome, 35-mm Kodachrome, and Professional 
Cut Sheet. The place of the meeting will be announced shortly. 

PACIFIC COAST SECTION 

At a meeting held on March 10th on the new scoring stage of the RCA Manu- 
facturing Co., at Hollywood, Mr. M. Rettinger, RCA Acoustical Engineer, pre- 
397 



398 SOCIETY ANNOUNCEMENTS 

sented a paper describing the changes made in the acoustical treatment of the 
stage for improved sound recording. Following this presentation, Mr. Olin 
Dupy of the Sound Department of Metro-Goldwyn-Mayer Studios discussed the 
use of mercury lamps in printers at the M-G-M Laboratory. The meeting 
closed with a presentation by Mr. Alston Rodgers of the Lamp Department of 
General Electric Company on "New Light-Sources in the Studios." His pres- 
entation was accompanied by a demonstration of the various light-sources. 



ADMISSIONS COMMITTEE 

At a recent meeting of the Admissions Committee, the following applicants for 
membership were admitted into the Society in the Associate grade: 

BAUM, WM. J. MAAG, G. A. 

War Dept. Theater No. 4 609 West 115th St., 

East Garrison, New York, N. Y. 

Camp Roberts, Calif. McCuLLOCH, J. S., JR. 

BREUNING,F. 36 Beaver St., 

5837 Camerford Ave., Sewickley, Pa. 

Hollywood, Calif. 

BUNCHEZ, S. PERRY ' B " R 

130 West 46th St., National Film Board, 

New York, N. Y. Ottawa, Canada 

FARLEY, W. L., JR. SMART, K. R. 

Eastman Kodak Company, National Carbon Co., Inc., 

6706 Santa Monica Blvd., 808 Olive St., 

Hollywood, Calif. St. Louis, Mo. 

JACOBSEN, CARL TRANSUE, WARREN 

4213 Beeman Ave., 6387 i vare ne Ave., 

North Hollywood, Calif. Hollywood, Calif. 
KALHUST, G. F. 

605 West Galena St., WARD ' A " A ' 

Butte, Montana 10459 Kinnaid Ave - 

KELSYE, R. M. Los An S eles ' Calif. 

Motion Picture Section, YOUNG, R. P. 

Engineer Board, 1214 Central St., 

Fort Belvoir, Va. Evanston, 111. 

In addition, the following applicants have been admitted to the Active grade : 

ANDREAS, J. M. COBB, L. L. 

120 North Madison Ave., 334 Huntley Drive, 

Pasadena, Calif. Los Angeles, Calif. 

SIMMONS, N. L., JR. 
Eastman Kodak Co., 

6706 Santa Monica Blvd., 
Hollywood, Calif. 



JOURNAL OF THE SOCIETY OF 
MOTION PICTURE ENGINEERS 

VOLUME XXXVIII MAY, 1942 

CONTENTS 



Recommended Practices of the Society of Motion 
Picture Engineers 403 

The Quarter- Wave Method of Speaker Testing 

S. L. REICHES 457 

Sound in Motion Pictures N. LEVINSON 468 

Current Literature . 483 

Fifty-first Semi-Annual Convention, Hollywood 
Hollywood, Calif., May 4-8, 1942 

General Information 484 

Abstracts of Papers 487 



{The Society is not responsible for statements of authors.) 



JOURNAL OF THE SOCIETY OF 
MOTION PICTURE ENGINEERS 

SYLVAN HARRIS, EDITOR 
Board of Editors 

ARTHUR C. DOWNES, Chairman 

JOHN I. CRABTREE ALFRED N. GOLDSMITH EDWARD W. KELLOGG 

CLYDE R. KEITH ALAN M. GUNDELFINGER CARLETON R. SAWYER 

ARTHUR C. HARDY 

Officers of the Society 

*President: EMERY HUSE, 

6706 Santa Monica Blvd., Hollywood, Calif. 
*Past-President: E. ALLAN WILLIFORD, 

30 E. 42nd St., New York, N. Y. 
*Executive Vice-President: HERBERT GRIFFIN, 

90 Gold St., New York, N. Y. 
** Engineering Vice-President: DONALD E. HYNDMAN, 

350 Madison Ave., New York. N. Y. 
*Editorial Vice-President: ARTHUR C. DOWNES, 

Box 6087, Cleveland, Ohio. 
** Financial Vice-President: ARTHURS. DICKINSON, 

28 W. 44th St., New York, N. Y. 
* Convention Vice-P resident: WILLIAM C. KUNZMANN, 

Box 6087, Cleveland, Ohio. 
^Secretary: PAUL J. LARSEN, 

1401 Sheridan St., N. W., Washington, D. C. 
* Treasurer: GEORGE FRIEDL, JR., 

90 Gold St., New York, N. Y. 

Governors 

*MAX C. BATSEL, 501 N. LaSalle St., Indianapolis, Ind. 
**FRANK E. CARLSON, Nela Park, Cleveland, Ohio. 

*JOHN G. FRAYNE, 6601 Romaine St., Hollywood, Calif. 

*ALFRED N. GOLDSMITH, 580 Fifth Ave., New York, N. Y. 
**EDWARD M. HONAN, 6601 Romaine St., Hollywood, Calif. 

*I. JACOBSEN, 177 N. State St., Chicago, 111. 
**JOHN A. MAURER, 117 E. 24th St., New York, N. Y. 

*LOREN L. RYDER, 5451 Marathon St., Hollywood, Calif. 

* Term expires December 31, 1942. 
** Term expires December 31, 1943. 



Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, included 
'in their annual membership dues; single copies, $1.00. A discount on subscription or single 
copies of 15 per cent is allowed to accredited agencies. Order from the Society of Motion 
Picture Engineers, Inc., 20th and Northampton Sts., Easton, Pa., or Hotel Pennsylvania, New 
York, N. Y. 

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 

General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 

Entered as second-class matter January 15, 1930, at the Post Office at Easton, 

Pa., under the Act of March 3. 1879. Copyrighted, 1942, by the Society of Motion 

Picture Engineers, Inc. 



RECOMMENDED PRACTICES 

OF THE 

SOCIETY OF MOTION PICTURE ENGINEERS 

The following pages contain all the recommended practices of the 
Society of Motion Picture Engineers now in force. Previous pub- 
lications of SMPE Standards and Recommended Practices appeared 
in the May 1930, November 1934, and March 1938 issues of the 
JOURNAL. 

Prior to a year or so ago specifications such as are contained in 
these pages were known as SMPE "Standards." By action of the 
Board of Governors the term "Standard" was discontinued, 
and only "Recommended Practices" are now issued by the Society. 
This terminology was adopted in order to avoid confusion with 
the standards approved by the American Standards Association. 
SMPE Recommended Practices when submitted to the American 
Standards Association through the Sectional Committee on Motion 
Pictures (Z-22) may become, when approved by the ASA, either 
American Standards or American Recommended Practices. 

The last publication of American Standards and American Recom- 
mended Practices appeared in the March 1941 issue of the JOURNAL. 
Not all SMPE Recommended Practices have been submitted to or 
approved by the ASA. In the following pages, those SMPE Recom- 
mended Practices that have been approved by the ASA bear the ASA 
Z-number in the upper left-hand corner of the page. The SMPE 
designation is given in the upper right-hand corner of each page. 
This designation includes a serial number and the year of adoption 
by the SMPE. Following the practice of the ASA, the serial num- 
bers will run in consecutive order and will never be used twice. 
That is, if, for example, the Recommended Practice SMPE-12-1936 
should be revised, the specification will be given a new number, such 
as SMPE-48-1943; if it should be rescinded, the number will merely 
be dropped. 

As this collection of SMPE Recommended Practices has been 
brought up to date, it may be understood that any specifications that 
appeared in previous issues of SMPE "Standards" have been super- 
seded, rescinded, or revised to one of the specifications on the follow- 
ing pages. 

403 



404 



RECOMMENDED PRACTICES 



[J. S. M. P. E. 



PAUL ARNOLD 
C. N. BATSEL 
M. C. BATSEL 
F. T. BOWDITCH 
F. E. CARLSON 
T. H. CARPENTER 
E. K. CARVER 

H. B. CUTHBERTSON 

L. W. DAVEE 
J. A. DUBRAY 



STANDARDS COMMITTEE 

D. B. JOY, Chairman 
A. F. EDOUART 
J. L. FORREST 
H. BARNETT 
A. N. GOLDSMITH 
H. GRIFFIN 
A. C. HARDY 
P. J. LARSEN 
C. L. LOOTENS 
J. A. MAURER 



G. S. MITCHELL 
K. F. MORGAN 
WM. H. OFFENHAUSER 
G. F. RACKETT 
W. B. RAYTON 

R. LlNDERMAN 

H. RUBIN 
O. SANDVIK 
H. E. WHITE 
E. W. TEMPLIN 



Reports of the Standards Committee 

(All references to J. Soc. Mot. Pict. Eng.} 
XV. (Aug., 1930), p. 160 

Safety Code for Projection; Wide-Film Dimensions. 
XV. (Dec., 1930), p. 818. 

Projector and Camera Speeds; Standard Release Print; Screen 

Brightness; Negative Notching; Wide-Film Dimensions. 
XVH. (Sept., 1931), p. 431. 

Wide-Film Dimensions. 
XVII. (Nov., 1931), p. 819. 

Glossary of Technical Terms Used in the Motion Picture In- 
dustry. 

XIX. (Nov., 1932), p. 477. 

16-Mm Standards. 

XX. (June, 1933), p. 505. 
XXII. (Jan., 1934), p. 17. 

Standard SMPE Film Perforation; Unit of Photographic In- 
tensity. 

Principle of Intermittency in Sensitometric Measurements. 
XXm. (Nov., 1934), p. 247. 

Standards Adopted by the SMPE. 

XXIV. (Jan., 1935), p. 16. 

Stresa (Italy) Conference on 16-Mm Standards. 

XXV. (July, 1935), p. 97. 

Report on Reichsfilmkammer at Berlin, April 25, 1935. 
XXV. (Aug., 1935), p. 192. 

XXV. (Oct., 1935), p. 370. 

International Standards Association Questionnaire Regarding 
16-Mm Sound-Film Standards. 

XXVI. (Jan., 1936), p. 18. 
XXVI. (May, 1936), p. 597. 

Great Britain Adopts SMPE 16-Mm Standards. 
XXVIII. (Jan., 1937), p. 21. 
XXVIII. (June, 1937), p. 585. 

16-Mm Reduction Printing. 



May, 1942] 



RECOMMENDED PRACTICES 



405 



XXIX. (Oct., 1937), p. 376. 

Report on Perforation Standards. 

XXX. (Mar., 1938), p. 248. 

Revision of SMPE Standards Proposed for Adoption by the 
Society. 

XXX. (Mar., 1938), p. 292. 

Projection Sprockets, Projection Reels, Sound Records and 
Scanned Area (35-Mm), Perforations. 

XXXI. (July, 1938), p. 65. 
XXXI. (Dec., 1938), p. 619. 

Cores for 35-Mm and 16-Mm Film, Sound-Track Dimensions, 
16-Mm Sound-Film Sprockets, Safety Film. 

XXXIV. (Jan., 1940), p. 88. 

Sound-Track Dimensions, Safety Film, 16-Mm Raw-Stock Cores. 

XXXV. (Nov., 1940), p. 525. 

Lantern-Slide Dimensions, Screen Brightness, Release-Print Sound- 
Track Specifications. 

XXXV. (Dec., 1940), p. 566. 

XXXVI. (Mar., 1941), p. 260. 

Raw-Stock Cores, Screen Brightness, Lantern-Slides, Raw-Stock 
Cutting and Perforating Specifications, Screen Brightness. 

XXXVII. (July, 1941), p. 76. 

XXXVII. (Nov., 1941), p. 535. 

Direction of Winding 16-Mm Film, Edge-Numbering 16-Mm 
Film. 

XXXVIII. (Jan., 1942), p. 87. 

Direction of Winding 16-Mm Film, Edge-Numbering 16-Mm 
Film. 



406 



RECOMMENDED PRACTICES 



[J. S. M. P. E. 



Same a 
ASA 
222.1 
1930 


SMPE RECOMMENDED PRACTICE 
t For 35-mm Motion Picture Film 


CUTTING AND PERFORATING 
NEGATIVE RAW STOCK* 








n 


o o 





c 
c 

c 
c 
-e 
c 


D C 

3 d 

\+ p 


D 
D 
D 

3-- 

r 

D 
D 

-_*~ 


? (- ") 


3 -e 


3 " 


3 6 C 

2_^^^-^^^^c 





Millimeters 


JncA Equivalents 


4 

B 
C** 
D 
E 
F 
G 
H 
LI 


35.00 + 0.00 
- 0.05 
4.75 0.013 
2.794 * 0.01 
1.85 0.01 
3.40 0.05 
28.17 0.05 
Not > 0.025 
2.08 0.025 
475.00 0.38 


1.378 + 0.000 
- 0.002 
0.1870 == 0.0005 
0.1100 0.0004 
0.0730 * 0.0004 
0.134 0.002 
1.109 =t 0.002 
Not > 0.001 
0.082 0.001 
18.70 0.015 


** Diameter of circle of curvature, 
t L length of any 100 consecutive perforation intervals. 



* For picture negative and certain special processes. 

These dimensions and tolerances apply to the material immediately 
after cutting and perforating. 

t Except that Z22. 11930 (see 7. Soc. Mot. Pic. Eng. t March, 1938, p. 261) 
was specified for both negative and positive raw stock. 



May, 1942] 



RECOMMENDED PRACTICES 



407 



SMPE RECOMMENDED PRACTICE 
For 35-mm Motion Picture Film 



CUTTING AND PERFORATING 
POSITIVE RAW STOCK* 



SMPE-2 
1941 




t 

D 

I 



C 





Millimeters 


Inch Equivalents 


A 

B 
C 
D 
E 
F 
G 
L** 
R 


35.00 + 0.00 
- 0.05 
4.750 * 0.013 
2.794 * 0.01 
1.98 0.01 
3.40 * 0.05 
28.17 =*= 0.05 
Not > 0.025 
475.00 * 0.38 
0.5 


1.378 + 0.000 

- 0.002 
0.1870 * 0.0005 
0.1100 * 0.0004 
0.0780 * 0.0004 
0.134 0.002 
1.109 * 0.002 
Not > 0.001 
18.70 * 0.015 
0.020 


** L = length of any 100 consecutive perforation intervals. 



* For positive prints and sound recording. 

These dimensions and tolerances apply to the material immediately 
after cutting and perforating. 



408 



RECOMMENDED PRACTICES 



[J. S. M. P. E. 







SMPE RECOMMENDED PRACTICE 
For 35-mm Motion Picture Sound-Film SMPE-3 




1936 
CAMERA APERTURE 




./^x- *"t 

a 


'^"~~ v ~l 

a 


, GUIDED EDGE 






a 


a 






a 




a 






cr 


J 






r 


LJ 


TRAVEL 




TT-* 




4- 








a 










-C 








a 


CAMERA 


\ ' 


i 










c 








^ 


rn 


APERTURE 








D 






I 






c 


D P 






~^ 


LJ 








c 




~~* 




*-6 


a 








1 


HK-E 






* 


p lm 








c 










a 


IMAGE 






c 


J 


II J 










c 

v*~ 




1 


a 




Dr 




1 

^ 


) 


4 OF CAMERA APERTURE 




4. OF FILM 





Millimeters 


Inch Equivalents 


A 
B 
C 
D 
E 
F 
G 
R 


22.05 0.05 
16.03 d= 0.05 
18.90 0.05 
2.97 
0.25 
3.07 
1.40 
0.8 approx. 


0.868 0.002 
0.631 * 0.002 
0.744 0.002 
0.117 
0.010 
0.121 
0.055 
0.03 approx. 


a = b = l /z longitudinal perforation pitch. 



The aperture dimensions given, in combination with the projector 
aperture shown in DS-35-4-1, result in a screen picture having a 
height-to-width ratio of 3 X 4 when the projection angle is 14 degrees. 

These dimensions and locations are shown relative to unshrunk 
raw stock. 



May, 1942] 



RECOMMENDED PRACTICES 



409 



SMPE RECOMMENDED PRACTICE 
For 35-mm Motion Picture Sound-Film 



PROJECTOR APERTURE 



SMPE-4 
1932 



i 


"a 
a 
a 
a 


N^\ ' 


^ 


^ ^ 

a 
a 

a 
a 


GUIDED EDGE 


TRAVEL 


1 
<p= 


) 





PROJECTOR 




a 


a 


* ' 






T A *" 


tc 


CD 
CD 

a 

CD 

a 

>F IMA 


B 

^^""'^ 

D 




a 


==i 


a 
a 
*"a 


APERTURE 
IMAGE 


-^ 


F- 


CD 










-J 

GE 


^\ 






- '^_ 

t OF HLM 


t . OF PROJECTOR APERTURE 





Millimeters 


JncA Equivalents 


A 


20.95 0.05 


0.825 0.002 


B 


15.25 0.05 


0.600 0.002 


C 


18.74 0.05 


0.738 0.002 


D 


0.39 


0.015 


E 


0.71 


0.028 


F 


0.38 


0.015 


G 


1.24 


0.049 


H 


0.15 


0.006 


R 


1 . 3 approx. 


. 05 approx. 


a = b l /z longitudinal perforation pitch. 



The aperture dimensions given result in a screen picture having 
a height-to-width ratio of 3 X 4 when the projection angle is 14 
degrees. 

These dimensions and locations are shown relative to unshrunk 
raw stock. 



410 



RECOMMENDED PRACTICES 



[J. S. M. P. E. 



Same as 
ASA 

Z22.2 
1941 



SMPE RECOMMENDED PRACTICE 
For 35-mm Motion Picture Sound-Film 



EMULSION AND SOUND RECORD 
POSITIONS IN CAMERA NEGATIVE 



SMPE-5 
1936 





f~^^ - 


// 


^ 


GUIDED EDGE 






% 













| 


a 



^^^LIGHT BEAM 
TRAVEL 







| 







\ 


a 
( } 


s 


a 
a 




: ;- i 




Q I] 1 -L 1 -L 

a 

a ,* 




a 
a 
a 





Drawing shows film as seen from inside the camera looking toward the 

camera lens. 



(1) Emulsion position in camera: toward the lens, except for 

special processes. 

(2) Speed : 24 frames per second. 

(3) Distance between center of picture and corresponding sound: 

20 frames. 



May, 1942] 



RECOMMENDED PRACTICES 



411 



Same as 
ASA 
Z22.3 
1941 


SMPE RECOMMENDED PRACTICE 
For 35-mm Motion Picture Sound-Film 


SMPE-6 
1936 


EMULSION AND SOUND RECORD 
POSITIONS IN PROJECTOR POSITIVE 
For Direct Front Projection 



GUIDED 




LIGHT BEAM 



Drawing shows film as seen from the light-source in the projector. 



(1) Emulsion position in projector: toward the light-source, except 

for special processes. 

(2) Speed: 24 frames per second. 

(5) Distance between center of picture and corresponding sound: 
20 frames. 



412 



RECOMMENDED PRACTICES 



U. S. M. P. E. 



Same as 


SMPE RECOMMENDED PRACTICE 




ASA 


For 35-mm Motion Picture Film 


SMPE-7 


Z22.4 




1938 




1941 


PROJECTION REELS 






-:r 



E: 



Capacity 


300 Meters 


1000 Feet 


600 Meters 


2000 Feet 


A 
B* 
C 


Millimeters 


Inch 
Equivalents 


Millimeters 


Inch 
Equivalents 


8.3Min 
40.1 
38.1 


0.328Min 
1.58 
1.50 


8.3Min 
40.1 
38.1 


0.328Min 
1.58 
1.50 


Recommended Practice 


r 

s 
t 


254.0 
3.17 
50.8 


10.0 
0.125 
2.0 


368.0 
3.17 
101.6 


14.5 
0.125 
4.0 



* This dimension applies only within a radius of 0.5 inch from the axis of the 
reel. 



May, 1942] 



RECOMMENDED PRACTICES 



413 



SMPE RECOMMENDED PRACTICE 
For 35-mm Motion Picture Film 



RAW-STOCK CORES 



SMPE-8 
1941 






Millimeters 


Inch 
Equivalents 


A 
B 
C 


25.90 =*= 0.20 
50.00 =*= 0.25 
34.50 0.50 


1.020 == 0.008 
1.968 0.010 
1.358 =*= 0.020 


Recommended Practice 


r 
s 


16.70 =*= 0.30 
4.00 0.20 


0.657 * 0.012 
0.157 0.008 



Bore ^4 to fit freely to hub 25.40 0.1 mm or 1.000 =*= 0.004 inch diameter. 



414 



RECOMMENDED PRACTICES 



tJ. S. M. i>. E. 



Same as 
ASA 
Z22.5 
1941 



SMPE RECOMMENDED PRACTICE 
For 16-mm Silent Motion Picture Film 



CUTTING AND PERFORATING 
NEGATIVE AND POSITIVE RAW STOCK 



SMPE-9 
1936 



CD 



CD 






Millimeters 


Inch Equivalents 


A 


16.00 + 0.00 


0.630 +0.000 




- 0.05 


- 0.002 


B 


7.620 0.013 


0.3000 0.0005 


C 


1.83 0.01 


0.0720 0.0004 


D 


1.27 0.01 


0.0500 =*= 0.0004 


E 


1.83 =*= 0.05 


0.072 0.002 


F 


12.320 0.025 


0.485 0.001 


G 


Not > 0.025 


Not > 0.001 


L* 


762.00 0.76 


30.0 =t 0.03 


R 


0.25 


0.010 


* L = the length of any 100 consecutive perforation intervals. 



These dimensions and tolerances apply to the material immediately 
after cutting and perforating. 



May, 1942] 



RECOMMENDED PRACTICES 



415 



Same as 
ASA 
Z22.6 
1941 


SMPE RECOMMENDED PRACTICE 
For 16-mm Motion Picture Film 


SMPE-10 
1934 


PROJECTOR SPROCKETS 







Number of Teeth in Mesh 


3 


4 


5 


6 




N 

6 

8 
12 
16 

6 

8 
12 
16 


B 


E 


C 


C 


C 


C 


(Holdback) ] Feed 


Mm 


In 


600' == 0.5' 
450 / 0.5' 
300' 0.5' 
2230' =* 0.5' 


Mm 


In 


Mm 


In 


Mm 


In 


Mm 


In 


14.38 
19.23 
28.93 
38.63 


0.566 
0.757 
1.139 
1.521 


1.02 
1.02 
1.02 
1.02 


0.040 
0.040 
0.040 
0.040 


0.91 
0.91 
0.91 


0.036 
0.036 
0.036 


0.79 
0.79 


0.031 
0.031 


0.69 
0.69 


0.027 
0.027 


14.15 
18.92 
28.50 
38.05 


0.557 
0.745 
1.122 
1.498 


600' 0.5' 
450' 0.5' 
300' 0.5' 
2230' 0.5' 


.02 
.02 
.02 
.02 


0.040 
0.040 
0.040 
0.040 


0.91 
0.91 
0.91 


0.036 
0.036 
0.036 


0.79 
0.79 


0.031 
0.031 


0.69 
0.69 


0.027 
0.027 


Combina- ] 
tion 


6 

8 
12 
16 


14.30 
19.13 

28.78 
38.43 


0.563 
0.753 
1.133 
1.513 


600' 0.5' 
450' 0.5' 
300' 0.5' 
2230' 0.5' 


.09 
.09 
.09 

1.09 


0.043 
0.043 
0.043 
0.043 


1.02 
1.02 
1.02 


0.040 
0.040 

0.040 


0.94 
0.94 


0.037 

0.037 


0.86 
0.86 


0.034 
0.034 


IFor All 
Sprockets 


A 
D 

r 
s 
t 
u 

V 


Millimeters 


Inch Equivalents 




Notes 


12.22 + 0.05 
-0.00 
1.22 + 0.00 
-0.08 
1.27 
0.08 
B-0.3, Max 
1.00 
B + 1.52. 
Max 


0.481 + 0.002 
- 0.000 
0.048 + 0.000 
-0.003 
0.050 
0.003 
B - 0.01, Max 
0.039 
B + 0.060, 
Max. 


2V = Number of teeth on sprocket. 
Tolerance for B and C +0.000 to 0.025 mm 
or +0.000 to -0.001 in. 
Dimensional standards indicated by capital 
letters. 
Recommended practice indicated by lower 
case letters. 
Values of C are omitted in cases where the 
angle of wrap on the sprocket would exceed 
180. 



416 



RECOMMENDED PRACTICES 



\J. S. M. P. E. 



Same as 
ASA 
Z22.7 
1941 


SMPE RECOMMENDED PRACTICE 
For 16 -mm Silent Motion Picture Film 


SMPE-11 
1936 


CAMERA APERTURE 




OF CAMERA APERTURE t. OF FILM 





Millimeters 


Inch Equivalents 


A 
B 
C 
D 
E 
F 
R 


10.41 0.05 
7.47 =*= 0.05 
8.00 0.05 
0.15 
0.05 
0.05 
0.5 approx. 


0.410 0.002 
0.294 * 0.002 
0.315 0.002 
0.006 
0.002 
0.002 
0.02 approx. 


a = b = l / 2 longitudinal perforation pitch. 



These dimensions and locations are shown relative to unshrunk 
raw stock. 

The center of the frame-line shall pass through the centers of per- 
forations on opposite sides of the film. 



May, 1942] 



RECOMMENDED PRACTICES 



417 



Same as 
ASA 
Z22.8 
1941 


SMPE RECOMMENDED PRACTICE 
For 16-mm Silent Motion Picture Film 


SMPE-12 
1936 


PROJECTOR APERTURE 




OF FILM 





Millimeters 


Inch Equivalents 


A 
B 
C 
D 
E 
F 
R 


9.65 =*= 0.05 
7.21 =*= 0.05 
8.00 0.05 
0.13 
0.38 
0.38 
0.5 approx. 


0.380 =*= 0.002 
0.284 0.002 
0.315 0.002 
0.005 
0.015 
0.015 
0.02 approx. 


a = b = l /z longitudinal perforation pitch. 



These dimensions and locations are shown relative to unshrunk 
raw stock. 

The center of the frame-line shall pass through the centers of 
perforations on opposite sides of the film. 



418 



RECOMMENDED PRACTICES 



LT. S. M. P. 



Same as 


SMPE RECOMMENDED PRACTICE 




ASA 


For 16-mm Silent Motion Picture Film 


SMPE-13 


79? a 




1Q9C 


1941 


EMULSION POSITION IN CAMERA NEGATIVE 


I99i 



a 



a 



LIGHT BEAM 
TRAVEL 



a 



Drawing shows film as seen from inside the camera looking toward the camera 

lens. 



(1) Emulsion position in camera: toward the lens, except for 

special processes. 

(2) Normal speed: 16 frames per second. 



May, 1942] 



RECOMMENDED PRACTICED 



419 



Same as 
ASA 
Z22.10 
1941 


SMPE RECOMMENDED PRACTICE 
For 16-mm Silent Motion Picture Film 


SMPE-14 
1936 


EMULSION POSITION IN PROJECTOR- 
POSITIVE 
For Direct Front Projection 



LIGHT BEAM 



a 



TRAVEL 



Drawing shows film as seen from the light-source in the projector. 



(1) Emulsion position in projector: toward the lens, except for 

special processes. 

(2) Normal speed: 16 frames per second. 



420 



RECOMMENDED PRACTICES 



[J. S. M. P. E. 



Same as 
ASA 
Z22.ll 
1941 


SMPE RECOMMENDED PRACTICE 
For 16- mm Motion Picture Film 


SMPE-15 
1938 


PROJECTION REELS 




A 
B 
C 


120 Meters 


400 Feet 


240 Meters 


800 Feet 


480 Feet 


1600 Feet 


Millimeters 


Inch Equivalents 


Millimeters 


Inch Equivalents 


Millimeters 


Inch Equivalents 


8.10 + 0.00 
-0.08 
8.10 + 0.00 
- 0.08 
17.2 Min 


0.319 + 0.000 
- 0.003 
0.319 + 0.000 
-0.003 
0.677 Min 


8.10 + 0.00 
-0.08 
8.10 + 0.00 
-0.08 
17.2 Min 


0.319 + 0.000 
-0.003 
0.319 + 0.000 
-0.003 
0.677 Min 


8.10 + 0.00 
- 0.08 
8.10 + 0.00 
-0.08 
17.2 Min 


0.319 + 0.000 
-0.003 
0.319 + 0.000 
-0.003 
0.677 Min 


Recommended Practice 


r 

5 

t 
u 

V 


177.8 
19.23 
37.85 
7.94 
3.17 


7.00 
0.757 
1.490 
0.312 
0.125 


250.8 
19.63 
37.85 
7.94 
3.17 


9.87 
0.773 
1.490 
0.312 
0.125 


355.6 
21.92 
117.48 
7.94 
3.17 


14.00 
0.863 
4.625 
0.312 
0.125 



NOTB: Center Spindle Holes. 
holes may be used. 



Either a combination of square and round holes or two square 



May, 1942] 



RECOMMENDED PRACTICES 



421 



Same as 

ASA 

Z22.12 

1941 



SMPE RECOMMENDED PRACTICE 
For 16-mm Motion Picture Sound-Film 



CUTTING AND PERFORATING 
NEGATIVE AND POSITIVE RAW STOCK 



SMPE-16 
1936 



B 





Millimeters 


Inch Equivalents 


A 

B 
C 
D 
E 
L* 
R 


16.00 +0.00 
- 0.05 
7.620 0.013 
1.83 0.01 
1.27 0.01 
1.83 0.05 
762.00 0.76 
0.25 


0.630 + 0.000 
- 0.002 
0.3000 0.0005 
0.0720 0.0004 
0.0500 0.0004 
0.072 0.002 
30.00 0.03 
0.010 


* L = the length of any 100 consecutive perforation intervals. 



These dimensions and tolerances apply to the material immediately 
after cutting and perforating. 



422 



RECOMMENDED PRACTICES 



U. S. M. P. E. 



Same as 
ASA 
Z22.13 
1941 


SMPE RECOMMENDED PRACTICE 
For 16-mm Motion Picture Sound-Film SMPE-17 

..... 1Q?A 


CAMERA APERTURE 


GUIDED EDGE. 


x-^ 





^ 




a 
o- 


TRAVEL 


a 

i 


CAMERA 
















\ 


b 




1 

L 1 


i 
I 


\ 






APERTURE 
IMAGE 


or CAN' 


~F 

^ . 

. 1 

ERA AF 


D 
-C - 


- 


KR 
r E 


a 

riLM 




L-^-x. 

>ERTURE 


3 

tor 





Millimeters 


Inch Equivalents 


A 
B 
C 
D 
E 
F 
R 


10.41 0.05 
7.47 0.05 
8.00 =*= 0.05 
0.15 
0.05 
2.79 
0.5 approx. 


0.410 0.002 
0.294 * 0.002 
0.315 == 0.002 
0.006 
0.002 
0.110 
0.02 approx. 


a = b = J /2 longitudinal perforation pitch. 



These dimensions and locations are shown relative to unshrunk 
raw stock. 

The center of the frame-line shall pass through the center of a 
perforation. 



May, 1942] 



RECOMMENDED PRACTICES 



423 



Same as 
ASA 
Z22.14 
1941 


SMPE RECOMMENDED PRACTICE 
For 16-mm Motion Picture Sound- Film 


SMPE-18 
1936 


PROJECTOR APERTURE 



GUIDED EDGE 






Millimeters 


Inch Equivalents 


A 
B 
C 
D 
E 
F 
R 


9.65 0.05 
7.21 0.05 
8.00 =t 0.05 
0.13 
0.38 
0.38 
0.5 approx. 


0.380 0.002 
0.284 0.002 
0.315 0.002 
0.005 
0.015 
0.015 
0.02 approx. 


a = b = l /z longitudinal perforation pitch. 



These dimensions and locations are shown relative to unshrunk 
raw stock. 

The center of the frame-line shall pass through the center of $ 
perforation. 



424 



RECOMMENDED PRACTICES 



tf. S. M. P. E. 



Same as 
ASA 
Z22.15 
1941 


SMPE RECOMMENDED PRACTICE 
For 16 -mm Motion Picture Sound- Film 


SMPE-19 
1936 


EMULSION AND SOUND RECORD 
POSITIONS IN CAMERA NEGATIVE 




Drawing shows film as seen from inside the camera looking toward the camera lens. 



(1) Emulsion position in camera: toward the lens, except for 

special processes. 

(2) Speed : 24 frames per second. 

(3) Distance between center of picture and corresponding sound: 

26 frames. 



May, 1942] 



RECOMMENDED PRACTICES 



425 



Same as 
ASA 

Z22.16 
1941 


SMPE RECOMMENDED PRACTICE 
For 16-mm Motion Picture Sound-Film 


SMPE-20 
1936 


EMULSION AND SOUND RECORD 
POSITIONS IN PROJECTOR POSITIVE 
For Direct Front Projection 



GUIDED EDGE 




LIGHT BEAM 



Drawing shows film as seen from the light-source in the projector. 



(1) Emulsion position in projector: toward the lens, except, for 

special processes. 

(2) Speed : 24 frames per second. 

(3) Distance between center of picture and corresponding sound: 

26 frames. 



426 



RECOMMENDED PRACTICES 



[J. S. M. P. E. 





SMPE RECOMMENDED PRACTICE 
For 16-mm Motion Picture Film 


SMPE-21 
1941 


EDGE-NUMBERING INTERVAL 



If 16-mm film is edge-numbered, the interval between consecutive 
footage numbers shall be 40 frames. 



May, 1942] 



RECOMMENDED PRACTICES 



427 





SMPE RECOMMENDED PRACTICE 
For 16-mm Motion Picture Film 


SMPE-22 
1941 


DESIGNATION OF DIRECTION OF WINDING 
OF FILM PERFORATED ALONG ONE EDGE 



When a roll of 16-mm film, perforated along one edge, is held so 
that the outside end of the film leaves the roll at the top and toward 
the right, winding A shall have the perforations on the edge of the film 
toward the observer; and winding B shall have the perforations on 
the edge away from the observer. In both cases the emulsion surface 
shall face inward on the roll. 

The following sketch illustrates these definitions: 




Winding A 
Emulsion side in 



Winding B 
Emulsion side in 



The above-given sketch shows film wound on cores. When the film is wound 
on a reel having a square hole on one side and a round hole on the other, the 
square hole shall be understood to be on the side away from the observer. 



428 



RECOMMENDED PRACTICES 



[J. S. M. P. E. 



SMPE RECOMMENDED PRACTICE 
For 16-mm Motion Picture Film 



RAW-STOCK CORES 



SMPE-23 
1941 






Millimeters 


Inch 
Equivalents 


A 
B 
C 


25.90 * 0.20 
50.00 * 0.25 
15.50 0.50 


1.020 0.008 
1.968 0.010 
0.610 =*= 0.020 


Recommended Practice 


r 
s 


16.70 0.30 
4.00 0.20 


0.657 * 0.012 
0.157 0.008 



Bore A to fit freely to hub 25.40 0.1 mm or 1.000 0.004 inch diameter, 



May, 1942] 



RECOMMENDED PRACTICES 



429 



Same as 
ASA 
Z22.17 
1941 


SMPE RECOMMENDED PRACTICE 
For 8-mm Motion Picture Film 


SMPE-24 
1938 


CUTTING AND PERFORATING 
NEGATIVE AND POSITIVE RAW STOCK 


SINGLE DOUBLE 
WIDTH WIDTH 








Inch 




Millimeters 


Equivalents 


A 


16.00 + 0.00 


0.630 + 0.000 




- 0.05 


- 0.002 


B 


3.810 0.013 


0.150 0.0005 


C 


1.83 0.01 


0.072 0.0004 


D 


1.27 * 0.01 


0.0500 0.0004 


E 


1.83 0.05 


0.072 * 0.002 


F 


12.320 * 0.025 


0.485 0.001 


G 


Not > 0.025 


Not > 0.001 


H 


8.00 + 0.00 


0.315 + 0.000 




- 0.08 


- 0.003 


L* 


381.00 * 0.38 


15.000 0.015 


R 


0.25 


0.010 


* L the length of any 100 consecutive perforation intervals. 



These dimensions and tolerances apply to the material immediately after cut- 
ting and perforating. 

Film may be slit before or after processing. 



430 



RECOMMENDED PRACTICES 



[J. S. M. P. E. 



Same as 

ASA 

Z22.18 

1941 



SMPE RECOMMENDED PRACTICE 
For 8-mm Motion Picture Film 



8-TOOTH PROJECTOR SPROCKETS 



SMPE-25 
1938 








Millimeters 


Inch 
Equivalents 


A 
B 

C 
D 
E 


5.72 0.03 
9.42 + 0.00 
- 0.05 
1.02 + 0.00 
- 0.05 
1.14 + 0.08 
- 0.00 
45 0' * 


0.225 0.001 
0.371 + 0.000 
- 0.002 
0.040 + 0.000 
- 0.002 
0.045 + 0.003 
- 0.000 
0.5' 


Recommended Practice 


r 
s 
t 
u 


2.54 
0.13 
0.51 
11.33 


0.100 
0.005 
0.020 
0.450 



May, 1942] 



RECOMMENDED PRACTICES 



431 



Same as 
ASA 
Z22.19 
1941 


SMPE RECOMMENDED PRACTICE 
For 8-mm Silent Motion Picture Film 


SMPE-26 
1938 


CAMERA APERTURE 



<fe OF FILM 



OF CAMERA APERTURE 






Millimeters 


Inch Equivalents 


A 
B 
C 
D 
E 
F 
R 


4.80 0.03 
3.51 0.03 
5.22 == 0.05 
0.30 
0.08 
0.76 
0.25 


0.189 0.001 
0.138 =*= 0.001 
0.205 0.002 
0.012 
0.003 
0.030 
0.010 


a = b = */2 longitudinal perforation pitch. 



432 



RECOMMENDED PRACTICES 



LT. S. M. P. E. 



Same as 
ASA 
Z22.20 
1941 


SMPE RECOMMENDED PRACTICE 
For 8-mm Silent Motion Picture Film 


SMPE-27 
1938 


PROJECTOR APERTURE 



OF FILM 




APERTURE 





Millimeters 


Inch 
Equivalents 


A 
B 
C 
D 
E 
F 
R 


4.37 0.03 
3.28 == 0.03 
5.22 == 0.05 
0.11 
0.21 
0.21 
0.25 


0.172 0.001 
0.129 0.001 
0.2055 0.002 
0.004 
0.008 
0.008 
0.010 


a = b = : /2 longitudinal perforation pitch. 



May, 1942] 



RECOMMENDED PRACTICES 



433 



Same as 


SMPE RECOMMENDED PRACTICE 




ASA 


For 8-mm Silent Motion Picture Film 


SMPE-28 


Z22.21 




1938 




1941 


EMULSION POSITION IN CAMERA NEGATIVE 





LIGHT BEAM 




Drawing shows film from inside the camera, looking toward the camera lens. 



(1) Emulsion position in camera: toward the lens, except for special 
processes. 

(2) Normal speed: 16 frames per second. 



434 



RECOMMENDED PRACTICES 



[J. S. M. p. E. 



Same as 
ASA 
Z22.22 
1941 


SMPE RECOMMENDED PRACTICE 
For 8-mm Silent Motion Picture Film 


SMPE-29 
1938 


EMULSION POSITION IN PROJECTOR POSITIVE 
For Direct Front Projection 



LIGHT BEAM 





TRAVEL 



Drawing shows film as seen from the light-source in the 
projector. 



(1) Emulsion position in projector: toward the lens, except for 
special processes. 

(2) Normal speed: 16 frames per second. 



May, 1942] 



RECOMMENDED PRACTICES 



435 



Same as 
ASA 
Z22.23 
1941 


SMPE RECOMMENDED PRACTICE 
For 8-mm Silent Motion Picture Film 


SMPE-30 
1938 


PROJECTION REELS 




Capacity, 60 M (200 Ft) 




Millimeters 


Inch Equivalents 


A 
B 


8.10 -h 0.00 
- 0.08 
8.9Min 


0.319 + 0.000 
- 0.003 
0.35Min 


Recommended Practice 


r 
S 
t 
U 

V 


127.0 
10.5 
37.8 
16.0 
1.6 


5.00 
0.41 
1.49 
0.63 
0.06 



Drive side of sprocket may have any desired odd number of driv- 
ing slots, evenly spaced. 



436 



RECOMMENDED PRACTICES 



Lf. S. M. P. E. 



SMPE RECOMMENDED PRACTICE 
For 35-mm Motion Picture Sound-Film 



FILM SPLICES 
NEGATIVE AND POSITIVE 



SMPE-31 
1936 





NEGATIVE SPLICE 




REGULAR POSITIVE SPLICE 




FULL HOLE POSITIVE SPLICE 





Negative 


Regular Positive 


Full Hole Positive 




Mm 


Inch Equiv. 


Mm 


Inch Equiv. 


Mm 


Inch Equiv. 


A 
B 
C 
D 
E 
F 


1.27 
4.75 
3.01 
3.01 
1.74 
1.74 


0.050 

0.187 
0.119 
0.119 
0.069 
0.069 


1.83 
4.75 
2.90 
3.68 
1.07 
1.85 


0.072 
0.187 
0.114 
0.145 
0.042 
0.073 


3.96 
9.50 
6.35 
7.11 
2.39 
3.15 


0.156 
0.374 
0.250 
0.280 
0.094 
0.124 



May, 1942] 



RECOMMENDED PRACTICES 



437 



Same as 

ASA 

Z22.24 

1941 



SMPE RECOMMENDED PRACTICE 
For 16-mm Silent Motion Picture Film 



FILM SPLICES 
NEGATIVE AND POSITIVE 



SMPE-32 
1936 




STRAIGHT SPLICE 





Diagonal 


Straight 




Mm 


Inch Equiv. 


Mm 


Inch Equiv. 


A 
B 
C 
D 
E 
F 


1.78 
7.62 
5.97 
3.53 
1.65 
4.09 


0.070 
0.300 
0.235 
0.139 
0.065 
0.161 


2.54 
15.24 
8.89 
8.89 
6.35 
6.35 


0.100 
0.600 
0.350 
0.350 
0.250 
0.250 



438 



RECOMMENDED PRACTICES 



[J. S. M. P. E. 



Same as 

ASA 

Z22.25 

1941 



SMPE RECOMMENDED PRACTICE 
For 16 -mm Motion Picture Sound -Film 



FILM SPLICES 
NEGATIVE AND POSITIVE 



SMPE-33 
1936 




STRAIGHT SPLICE 





Diagonal 


Straight 




Mm 


Inch Equiv. 


Mm 


Inch Equiv. 


A 
B 

D 
E 
F 


1.78 
7.62 
5.97 
3.53 
1.65 
4.09 


0.070 
0.300 
0.235 
0.139 
0.065 
0.161 


2.54 
15.24 
8.89 
8.89 
6.35 
6.35 


0.100 
0.600 
0.350 
0.350 
0.250 
0.250 



May, 1942] 



RECOMMENDED PRACTICES 



439 



Same as 
ASA 
Z22.26 
1941 


SMPE RECOMMENDED PRACTICE 
For Motion Picture Film 


SMPE-34 
1934 


SENSITOMETRY 



The principle of non-intermittency shall be adopted as recom- 
mended practice in making sensitometric measurements. 



440 



RECOMMENDED PRACTICES 



[J. S. M. P. E. 



Same as 
ASA 
T2221 
1941 


SMPE RECOMMENDED PRACTICE 
For Motion Picture Film 


SMPE-35 
1938 


PHOTOGRAPHIC DENSITY 



The integrating sphere shall be used as a primary instrument for 
the determination of photographic density. Photographic densities 
determined by means of this primary instrument shall be used as 
secondary or reference standards by means of which densitometers 
of other types may be calibrated. 



May, 1942] RECOMMENDED PRACTICES 441 



Same as 
ASA 
Z22.28 
1941 


SMPE RECOMMENDED PRACTICE 
For Motion Picture Film 


SMPE-36 
1938 


PROJECTION ROOMS 



Projection Lens Height. The standard height from the floor to 
the center of the projection lens of a motion picture projector should 
be 48 inches. 

Projection Angle. Should not exceed 12 degrees. 

Observation Port. Should be 12 inches wide and 14 inches high, 
and the distance from the floor to the bottom of the openings shall 
be 48 inches. The bottom of the opening should be splayed 15 de- 
grees downward. If the thickness of the projection room wall 
should exceed 12 inches, each side should be splayed 15 degrees. 

Projection Lens Mounting. The projection lens should be so 
mounted that the light from all parts of the aperture shall traverse 
an uninterrupted part of the entire surface of the lens. 

Projection Lens Focal Length. The focal length of motion picture 
projection lenses should increase in y 4 -inch steps up to 8 inches, 
and in 1 / 2 -inch steps from 8 to 9 inches. 

Projection Objectives, Focal Markings. Projection objectives should 
have the equivalent focal length marked thereon in inches, quarters, 
and halves of an inch, or in decimals, with a plus (+) or minus ( ) 
tolerance not to exceed 1 per cent of the designated focal length also 
marked by proper sign following the figure. 



(Complete plans for projection rooms are contained in J. Soc. Mot. Pict. Eng., 
Nov., 1938, p. 484.) 



442 



RECOMMENDED PRACTICES 



[J. S. M. P. E. 



Same as 
ASA 
122.29 
1941 


SMPE RECOMMENDED PRACTICE 
For 35-mm Motion Picture Film 


SMPE-37 
1938 


PROJECTION SCREEN DIMENSIONS 



Sizes of screens shall be in accordance with the table below. 

The spacing of grommets shall be 6 inches, with 12 inches as a 
possible sub-standard. The ratio of width to height of screens shall 
be 4 to 3. 

The width of the screen should be equal to approximately Ye 
the distance from the screen to the rear seats of the auditorium. 
The distance between the front row of seats and the screen should 
be not less than 0.87 foot for each foot of screen width. 



Screen Sizes 



Size No. 
of Screen 


Picture 
Width 
(Feet) 


Picture 
Feet 


Height, 
Inches 


Size No. 
of Screen 


Picture 
Width 
(Feet) 


Picture 
Feet 


Height, 
Inches 


8 


8 


6 





25 


25 


18 


9 


9 


9 


6 


9 


26 ' 


26 


19 


6 


10 


10 


7 


6 


27 


27 


20 


3 


11 


11 


8 


3 


28 


28 


21 





12 


12 


9 





29 


29 


21 


9 


13 


13 


9 


9 


30 


30 


22 


6 


14 


14 


10 


6 


31 


31 


23 


3 


15 


15 


11 


3 


32 


32 


24 





16 


16 


12 





33 


33 


24 


9 


17 


17 


12 


9 


34 


34 


25 


6 


18 


18 


13 


6 


35 


35 


26 


3 


19 


19 


14 


3 


36 


36 


27 





20 


20 


15 





37 


37 


27 


9 


21 


21 


15 


9 


38 


38 


28 


6 


22 


22 


16 


6 


39 


39 


29 


3 


23 


23 


17 


3 


40 


40 


30 





24 


24 


18 














May, 1942] 



RECOMMENDED PRACTICES 



443 





SMPE RECOMMENDED PRACTICE 
For 35-mm Motion Picture Film 


SMPE-38 
1941 


PROJECTION SCREEN BRIGHTNESS 



The brightness at the center of a screen for viewing 35-mm motion 
pictures shall be lOJtt ft-lamberts when the projector is running, 
with no film in the gate. 



444 



RECOMMENDED PRACTICES 



[J. S. M. P. E. 





SMPE RECOMMENDED PRACTICE 
For 35-mm Motion Picture Sound-Film 


SMPE-39 
1938 


SOUND TRANSMISSION OF SCREENS 



A loss of 2.5 db as given by the average response curve at 6000 cps 
relative to the 1000-cycle response as recorded, is a desirable limiting 
value for existing types of sound equipment. Screens that meet this 
requirement are usually found to attenuate 4 db at 10,000 cps. 
As to regularity of response, variations greater than 2 db would not 
be tolerable. No limits for regularity have been established for 
frequencies lower than 300 cps. 



May, 1942] 



RECOMMENDED PRACTICES 



445 



Same as 
ASA 
Z22.30 
1941 


SMPE RECOMMENDED PRACTICE 
Definition for Motion Picture Apparatus 


SMPE-40 
1930 


NUMBER OF TEETH IN MESH 



The number of teeth in mesh with the film (commonly referred to as 
"teeth in contact") shall be the number of teeth in the arc of contact 
of the film with the drum of the sprocket when the pulling face of 
one tooth is at one end of the arc. 



446 RECOMMENDED PRACTICES [J. S. M. P. E. 



Same as 
ASA 
Z22.31 
1941 


SMPE RECOMMENDED PRACTICE 
For Motion Picture Film 


SMPE-41 
1934 


SAFETY FILM 



The term "Safety Film," as applied to motion picture materials, 
shall refer to materials having a burning time greater than 10 seconds 
and falling into the following classes : (a) support coated with emul- 
sion, (b) any other material upon which or in which an image can be 
produced, (c) the processed products of these materials, and (d) 
uncoated support that is or can be used for motion picture purposes 
in conjunction with the aforementioned classes of materials. 

The burning time is defined as the time in seconds required for the 
complete combustion of a sample of the material 36 inches long, the 
determination being according to the procedure of the Underwriters 
Laboratory. This definition was designed specifically to define Safety 
Film in terms of the burning rate of the commercial product of any 
thickness or width used in practice. The test of burning time, there- 
fore, shall be made with a sample of the material in question having 
a thickness and width at which the particular material is used in 
practice. 

All 16 and 8-mm film must be of the safety type. 



May, 1942] RECOMMENDED PRACTICES 447 



Same as 


SMPE RECOMMENDED PRACTICE 




ASA 

799 79 


For 35-mm Motion Picture Sound-Film 


SMPE-42 

1 Q4A 


LLL.SL 

1941 


FADER SETTING INSTRUCTIONS 


19tv 



The Fader Setting Instruction Leader shall consist of 15 frames 
located in the first 20 frames of the synchronizing leader; the first 
frame shall designate the type of print; the second frame the type 
of reproducing equipment necessary to project the print; and the 
next nine frames the general fader setting specified in relation to an 
average fader setting for the particular product under consideration. 
The remaining frames may be used for whatever additional informa- 
tion the studio may wish to transmit to the theater. 

The designation "Regular" in the Instruction Leader indicates 
that only one type of print has been issued on the particular produc- 
tion under consideration. Productions with prints designated as 
either "Hi-Range" or "Lo-Range" are issued in both types of prints, 
i. e., all productions on "Hi-Range" prints will have necessarily been 
issued on "Lo-Range" prints as well. 

Both the terms "push-pull" and "single" shall be on every leader, 
one or the other being crossed out to leave the proper term desig- 
nating the type of sound-track on the print. 



448 



RECOMMENDED PRACTICES 



LT. S. M. P. E. 




May, 1942] 



RECOMMENDED PRACTICES 



449 



Same as 


SMPE RECOMMENDED PRACTICE 




ASA 


For 35-mm Motion Picture Sound-Film 


SMPE-43 


122.33 
1941 


FILTER NOMENCLATURE 


1WV 



The symbol describing any filter shall consist of three characters, 
the first designating the frequency of 3-db insertion loss, the second 
the character "Hi" or "Lo" to indicate high-pass or low-pass, and 
the third the frequency of 10-db insertion loss (all frequencies in 
cycles). 

Thus the following describes several low-pass filters "4000 Lo 
6000" (Fig. 1), "5000 Lo 7000" or "4500 Lo 5500" and the following 
describe several high-pass filters: "60 Hi 40" (Fig. 2), "80 Hi 30," 
or "100 Hi 50." 

A combination of two of the above symbols may be used to de- 
scribe a band-pass filter (Fig. 3) or a dividing network (Fig. 4) or a 
reverse combination of symbols may be used to describe a band- 
elimination filter (Fig. 5). 



450 



RECOMMENDED PRACTICES 



[J. S. M. P. E. 





















1 

0<X 


II 

-L0-000 



































S 






8 


























^ 






























\ 


s 






























' 












nVQUCNC 


r 


















fICUK 1-LOW-MkM rt.Tf.lt 












0-M-0 





























^ 






















8 






/ 




























/ 




























/ 
































































I* 


1 


flOOBC 2-M-M5 riLTtH 


1000 M 


X) 


4O 


00 


wooo 












II 

TO-MI-SO <>-a>-100 























^~ 










^ 












8 








/ 










\ 


















/ 












\ 
















/ 




































nirnvur 





























1 1 

00-l 


IN H 










1 1 

*00-MI-ZOO 








- 














> 


/ 
















8 












/ 






x 
























X 










\ 






















/ 










\ 












































* " *o 


* ICUMC *-WVlD.NC 


rwof 




00 


*** 









1 


JOO-LO-400 














JO 


M 


M - 


il 













s^ 










J 











8 












\ 








/ 
























\ 




/ 




























\ 


/ 




















.^.fsa^s 


2 






















3 

8 
>> 

i 

o 

< 

I 

s 



rt 



I 

t 






s 

C0 




o p 

03 ta 

fi 

*rt -M 



e 



.^ 
T3 PH 

I 
II 

2 

J 1 

i ? 



* 

CO 

c 


i 



re 



- 
s 



55 



it 





1.1- 
- 






ifl 

Si 

' 



0-s 



C +3 






6 
di 
ct 



s 



e 
e 



he frames in which the numerals 
inverted) shall be placed imm 
bility of mis-reading in the proj 
verted numerals. 



O ' 

--t- ) 









S -M Z <J 

c 3 a J 
c o S 'T; 
o 3 ^ o - 

2 -M OJ=1 o 



"o rt a3 rt .2 

a^ e^-o 

^ !l 



the first fram 
ame containi 
tage indicator 
point exactly 
there shall b 
inch wide. 



the 
ra 
tag 
At a p 
fr t 
b 



a 
e 



from 
rent f 
. Foo 

At 
fram 
h by 3 / 




ning 3 feet 
a transpa 
me height. 
inclusive. 
ge numeral f 
s inch high 












-g 






-3 






pictur 
ifican 
he pic 
betw 
eet n 



Ills 




v 

1-8 



l 
s 

R 



5 v- --^ 

e o"S ^ 
o^ S a- 2 8 

.2 6 6 S 
^^o^ 2 2 
o* .S c 

rt' c! ' aJH o'aJ 

3^.S 3 ^ 
S.^^^ a^ 

.fa A 03 W ^ -9 -M 1 
J kO-^3 ^ <L 

+J 4^ D 

a; a; O bf^ 

^.MSwf fl 

^3 ti'O a; bo -Hi 

i C ct 



he 
d 
y 
an 



from th 
serrate 
halfwa 
right-h 
server 
there 






* Sf 

' 



be 



May, 1942] RECOMMENDED PRACTICES 453 



INFORMATIONAL DATA 
For Motion Picture Film 



UNIT OF PHOTOGRAPHIC INTENSITY 



1941 



While no standard or recommended practice for a unit of photo- 
graphic intensity is proposed at this time* the description of an 
international unit of photography as adopted by the International 
Congress of Photography in 1928 and 1931 is here given as a matter 
of information. The resolution passed by the 1928 Congress defining 
the unit of photographic intensity received the approval of the 
English and American national committees and of the Optical 
Society of America. At the 1931 Congress an amendment to the 
resolution passed by the 1928 Congress was proposed and accepted 
by the representatives of the various national committees in at- 
tendance. So far as can be determined, no official approval by the 
various national committees was subsequently given. There is 
therefore a little doubt as to the exact status of this standard as 
established by the International Congress. The ASA Sectional 
Committee Z22, Motion Pictures, is at the present time in active 
cooperation with the ASA Sectional Committee Z38, Photography, 
in an endeavor to clarify this matter and to formulate a proposal 
for an American standard for a unit of photographic intensity. 

The unit of photographic intensity for the sensitometry of nega- 
tive materials may be defined as the intensity of a filtered source of 
radiation having a luminous intensity of one international candle, 
and produced by a gray body at a color temperature of 2380 (accord- 
ing to the most recent determination of the international tempera- 
ture scale), together with a selectively absorbing filter made up as 
follows: Two solutions compounded according to the following 
formula, the complete filter to consist of a 1-cm** layer of each solu- 
tion contained in a double cell made by using three pieces of boro- 
silicate crown glass (refractive index, D = 1.51), each 2.5 mm thick. 

* Reprinted from J. Soc. Mot. Pict. Eng. (March, 1941), p. 256. 
** Tolerance in thickness shall be 0.05 mm. 



454 RECOMMENDED PRACTICES [J. S. M. P. E. 

Solution A 

Copper sulfate (CuSO 4 , 5H 2 O) 3.707 gm* 

Mannite (CeH 8 (OH) 6 ) 3 . 707 gm* 

Pyridine (C 6 H 5 N) 30.0 cc 

Water (distilled) to make 1000. cc 

Solution B 
Cobalt ammonium sulfate 

(CoS0 4 , (NH 4 ) 2 S0 4 , 6H 2 0) 26.827 gm* 

Copper sulfate (CuSO 4 , 5H 2 O) 27. 180 gm* 

Sulfuric acid (sp. gr. 1.835) 10.0 cc 

Water (distilled) to make 1000. cc 

The spectrophotometric absorption characteristics of the filter 
made up according to these specifications are shown in the following 
chart,** and in the table below this information is given in numerical 
form. 

The unit of photographic intensity as recommended by the Seventh Inter- 
national Congress of Photography, held in London in July, 1928, was formally 
approved by the English and American Committees of the Congress and by the 
Optical Society of America. In ratifying this standard, the Optical Society of 
America, in order to forestall any possible misinterpretation of the intent of the 
resolution, presented the following clarifying statements: 

"In ratifying this proposal the Optical Society of America understands that the 
intent of this recommendation is as follows: 

(1) The intention is to specify two things, (a) the unit hi which the intensities 
of light-sources are to be expressed, and (6) the quality of light to be used. 

(2) The unit is to be the International Candle, implying further that the in- 
tensities measured and stated will be luminous intensities as in visual photometry. 

(5) The quality of light to be used for sensitometry of negative materials is to 
be that which results from passing the radiation from a gray body at 2380 K nor- 
mally through the filter described. 

(4) The gray body and the selectively absorbing filter together shall be con- 
sidered as the effective source in specifying the intensity (candle-power). 

* For practical purposes an accuracy to the second place of decimals is prob- 
ably sufficient. 

** Bur. Standards Misc. Publ. No. 114: Raymond Davis and K. S. Gibson, 
"Filters for the Reproduction of Sunlight and Daylight, and the Determination 
of Color-Temperature." 



May, 1942] 



RECOMMENDED PRACTICES 



455 



ISO 



125 



100 



75 



50 



25 



.50 



50 400 450 500 550 60O 650 700 756 

Wavelength millimicrons (m/i) 



X 

(m M ) 


T 


E"* 


E"/E' 


(DIM) 


T 


E"* 


E"/E' 


350 


0.230 


2.5 




40 


0.182 


100.6 


0.997 


60 


0.321 


4.9 


0.304 


550 


0.167 


103.2 


1.014 


70 


0.420 


8.7 


0.426 


60 


0.1495 


103.1 


1.031 


80 


0.518 


14.5 


0.579 


70 


0.1316 


100.4 


1.021 


90 


0.608 


22.5 


0.749 


80 


0.1151 


96.9 


0.997 


400 


0.671 


32.3 


0.715 


90 


0.1013 


93.4 


0.977 


10 


0.714 


44.0 


0.767 


600 


0.0907 


91.3 


0.959 


20 


0.737 


57.4 


0.871 


10 


0.0834 


91.2 


0.968 


30 


0.725 


70.3 


1.016 


20 


0.0778 


92.3 


0.991 


40 


0.682 


81.2 


1.044 


30 


0.0730 


93.3 


1.012 


450 


0.612 


88.6 


1.021 


40 


0.0684 


94.0 


1.032 


60 


0.540 


93.9 


1.018 


650 


0.0646 


95.1 


1.060 


70 


0.483 


99.9 


1.031 


60 


0.0605 


95.1 


1.074 


80 


0.431 


105.0 


1.060 


70 


0.0558 


93.4 


1.081 


90 


0.376 


107.1 


1.065 


80 


0.0510 


90.5 


1.069 


500 


0.315 


103.9 


1.020 


90 


0.0457 


85.8 


1.038 


10 


0.259 


98.3 


0.972 


700 


0.0409 


81.2 


1.008 


20 


0.219 


95.0 


0.939 


10 


0.0366 


76.4 


0.979 


30 


0.197 


96.7 


0.957 


20 


0.0326 


71.6 


0.940 



T Spectral Transmission of Filter at 

25C 

V Relative Visibility Function 
E Relative Energy of 2360K 
E' Relative Energy of Mean Noon 

Sunlight at Washington 



E ff ( = T X )* Relative Energy of 
2380 K and Filter Combination 

Light Transmission of Filter 
for 2380K = 0.1352** 



* Adjusted to make sum of E" E' from 400 to 720 m/* equal practically to 



zero. 



** Factor to be used to multiply the candle-power of the light-source to obtain 
the candle-power of the source-and-filter combination. 



456 RECOMMENDED PRACTICES 

(5) The procedure recommended for determining the intensity of the com- 
bined effective source is to multiply the intensity of the primary source (gray 
body) by the appropriate transmission factor of the filter which is 0.135. This 
factor has been computed from the spectral transmission of the filter via the rela- 
tive energy distribution of 2380 K and the relative visibility function adopted by 
the sixth session of the International Commission on Illumination at Geneva, 
1924. 

(6) This resolution does not state or imply the value of illumination to be used 
at the test plane during the sensitometric exposure, nor does it place any limita- 
tions on the intensity of the light-source to be used. 

Pyridine as included in Solution A can, in the opinion of various scientific 
groups which have investigated the matter, be obtained commercially of sufficient 
purity to obviate the need for any elaborate precautions and to permit its satis- 
factory inclusion in the formula for Solution A." 



THE QUARTER- WAVE METHOD OF SPEAKER TESTING* 

S. L. REICHES** 



Summary. The theory and use of a method for testing loud speakers in theaters 
are described and presented as a tool for the theater serviceman. Using this method, 
the serviceman can quickly, and with an accuracy commensurate with his require- 
ments, determine the average frequency response of the speakers, as seen from various 
points in an auditorium. He can determine also, for a specific auditorium, the horn- 
system distribution pattern that allows the use of a minimum of acoustic treatment 
and also the position of this treatment. 

A simple method of loud speaker testing will be described, along 
with the results of experiments showing its usefulness. This method 
will be called "the quarter- wave method," since the driving force 
employs what amounts to a quarter of a sine-wave generated as an 
exponential voltage pulse. The output of the speaker is measured 
by amplifying the signal picked up by a microphone, and reading the 
peak voltage obtained. By this method it is possible to test speaker 
response vs. frequency at various positions in a sound-field when 
several speakers are employed, and to test the phasing of one speaker 
relative to the others. Since a single pulse is employed, and only 
the peak voltage of the pick-up is used, it is possible to test a speaker 
or set of speakers in their natural locations without interference by 
echoes or reverberation due to nearby walls. 

Application of the Quarter-Wave Pulse. The quarter-wave ex- 
ponential voltage pulse is generated by the discharge of condenser 
C through resistance R in the circuit shown in Fig. 1. This pulse is 
sent through the amplifier system and applied to the speaker or 
speakers to be tested. It is true that the pressure pulse communi- 
cated to the air by this method can not be strictly represented by an 
expression of the type p = poe~ at . In actual practice, however, 
the pressure pulse rises sharply to a maximum, then drops off in a 
more or less exponential fashion. This has been demonstrated by 
connecting the pick-up to a cathode-ray oscilloscope. 

* Presented at the 1941 Fall Meeting at New York, N. Y. ; received January 
22, 1942. 

** The Brush Development Co., Cleveland, Ohio. 

457 



458 



S. L. REICHES 



LT. S. M. P. E. 



The correlation of the quarter-wave exponential pulse with fre- 
quency is accomplished as follows: Consider a voltage 



and 



V = V*r<* = Ve-RC (t > 0) 
V = ( 0) 




CHARGE 
DISCHARGE 


sj 

T 


ii 

Oui 



B+ 
FIG. 1. Quarter- wave generator. 




FIG. 2. Distribution curve of outside horns. 

applied to a speaker cone. This gives rise to a current / = ( V /Z)e- at 
where Z is the impedance of the speaker. The force on this cone at 
t = will be proportional to a(Vo/Z) (the time rate of change of the 
voltage) and for t > 0, it will diminish by the factor e-**. Since it 
is known that the pressure communicated to the air by the speaker 
cone rises sharply to a maximum, we may say without great error 
that the peak pressure in the air is determined by the force a(Vo/Z). 
On the other hand, if we had applied a simple harmonic wave V = 



May, 1942] QUARTER- WAVE SPEAKER TESTING ' 459 

Fo sin ut to the speaker, by the same argument as before, the maxi- 
mum force on the speaker cone would be coFo/Z. If we make w = 
a = 2ir/, where / is the steady-state sine- wave frequency, we should 
expect the same peak pressures in the sound pulses obtained by the 
two methods. We may then say 



2ir 2<rRC 

and we have a correlation between the exponential and the sinusoidal 
excitations. 

The Method of Testing. The quarter-wave generator is connected 
to the input of the system to be tested. A schematic diagram of 
such a device is shown in Fig. 1. A pressure actuated microphone is 
placed in the sound-field at whatever point desired to make a test. 
The generator is actuated, and readings taken by reading a peak 
voltmeter connected to the microphone. Readings are taken for the 
various R-C combinations that give the frequencies at which it is 
desired to make tests. The usual time required to make one set of 
ten readings is about five minutes; thus tests may be conducted at 
several points in a room in a comparatively short time. 

Tests in the Cleveland Stadium Using the Quarter-Wave Technic. 
The public address installation in the Cleveland Stadium pre- 
sented some interesting problems. For a number of reasons, none 
connected with engineering procedure, it was necessary to place the 
horns in a cluster at the foot of the bleachers ; these horns are Western 
Electric Type 598. At the time this study was begun, the faults 
were: 

(1) Insufficient power, even with a partially empty stadium, with 2000 watts 
of audio available. 

(2) Low intelligibility due to as many as ten repetitions of the same sound 
reflected from various parts of the stadium. 

(5) An effect that might be described as a "rolling-around" of sound between 
the upper tier of seats and the main floor. 

Determination of Improper Phasing. A pulse equivalent to 350 
cycles was chosen as a study frequency. Two considerations suggest 
the use of 350 cycles first, the maximum energy of speech occurs 
around this frequency; second, other factors such as spacing, etc., 
become equal in importance to the electrical problems at higher fre- 
quencies. 



460 



S. L. REICHES 



Q. S. M. P. E. 




FIG. 3 (a). Distribution curve for inside horns. 




FIG. 3 (&. Distribution curve of all horns. 




FIG. 4. Comparison of theoretical and actual distri- 
bution curves of initial installation. 



May, 1942] QUARTER- WAVE SPEAKER TESTING 461 

The distance from the horns for readings was 300 feet. This dis- 
tance was considered sufficient to insure the complete formation of 
the beam from all four speakers. 

The desired procedure consisted of measuring the distribution pat- 
tern of each horn, and by combining these patterns, getting the 
maximal output in each direction from the combination, then making 
a comparison between this and the measured combined distribution 
pattern. 

It was not possible to measure each speaker in the stadium sepa- 
rately, as the speakers are driven in pairs by separate power ampli- 
fiers, the outside speakers comprising one pair, and the center two 
another pair. This made it necessary to measure the outside and 
inside speakers in pairs. These measurements are plotted on Fig. 2 
for the outside pair, and on Fig. 3 (a) for the inside pair. From these 
patterns it is evident that the speakers comprising each pair are 
quite well phased. Much interference between the outer speakers 
would not be expected, even if they were out of phase, since they 
were pointed away from each other, and were separated quite widely. 

Fig. 3(b) shows the distribution for all four horns. 

Fig. 4 shows the difference between the actual measured curve and 
the theoretical curve determined by adding the distribution pattern 
of each pair. The power loss is quite large. 

The offending speaker, which was one of the outside speakers, 
was found to have a transposed field coil. Fig. 5 shows a comparison 
between the new measured distribution pattern and a new theoretical 
pattern. (This new theoretical pattern was produced by redirection 
of one of the speakers.) 

The improvement was quite evident. Of interest is the fact that 
on the same afternoon of the correction only about 500 watts were 
required to produce the desired loudness with some 25,000 persons 
present. 

Conclusions on Phasing. Probably this correction could have been 
made without taking the data shown here. However, there are some 
interesting points shown by the patterns of Figs. 3 to 5. The most 
important of these can be seen from Fig. 5. This shows that al- 
though all speakers are properly connected, it is not possible to 
realize the full theoretical power output of all speakers due to their 
physical spacing and possibly to the variation in electrical phase- 
shift in the driving amplifiers. There is still much interference. 

The Optimal Distribution Pattern. Every area is associated with a 



462 



S. L. REICHES 



[J. S. M. P. E. 




FIG. 5. Comparison of theoretical and actual dis- 
tribution after rephasing. 




FIG. 6. Comparison of optimum curve with actual 
curve. 




FIG. 7. Comparison of optimum curve with final 
curve. 



May, 1942] 



QUARTER- WAVE SPEAKER TESTING 



463 



no 



FREQUENCY 



FIG. 8. Frequency characteristics of entire horn system. 



0*5 



s ir t . 

FREQUENCY 

FIG. 9. Frequency characteristics of good sound quality 
house. 

distribution pattern that will produce equal loudness to an observer 
at all points on the periphery of this area. The desirability of such 
a distribution is obvious from the standpoint of good listening. This 
distribution pattern is called "the optimal distribution pattern," 
that is, the sound at any point on the boundary shall not be too loud 
for one observer or too soft for another. In addition to this, the 
optimal distribution pattern will produce the least amount of re- 
flected sound, and thereby the greatest intelligibility. This can be 
seen from Fig. 6, which shows the comparison between the measured 



464 



S. L. REICHES 



Lf. S. M. P. E. 



CENTER 



REARCE 



\ FREQUENCY 

FIG. 10 (a). Frequency characteristics of poor sound quality 
hotlse. 




FIG. 10 (b). Final distribution curve. 

distribution of all four horns, and the optimal distribution pattern 
for an area the shape of the Cleveland Stadium. The cardioid- 
shaped optimal distribution pattern was found from the relation- 
ship: 



where 

DB l 
DB 2 



DB 2 = DB l - 20 log 5r 



desired level which occurs at greatest distance di from horns, 
unknown level at shorter distance d z . 



It will be seen from this that there is a large excess of energy 
directly down the center of the stadium. This will cause the loud- 
ness to be too great in this area, and the reflected sound will be much 



May, 1942] QUARTER- WAVE SPEAKER TESTING 465 

greater than necessary. It follows that were acoustic treatment to 
be applied, the greatest amount would have to be placed in this area. 

Fig. 7 shows the final corrected pattern, in which there is some 
variation from the optimum. The extra level between 10 and 60 
degrees was introduced to compensate for the prevailing winds from 
Lake Erie, the data included here being taken on quiet mornings with 
no wind. 

Conclusions on Optimal Distribution Pattern. The improvement 
obtained was quite pronounced even to untrained observers. With 
the resultant improvement in intelligibility, the apparent multi- 
plicity of source was completely eliminated. This improvement in 
intelligibility allowed a further decrease in driving power. 

It was hoped that some improvement would also be made in regard 
to the "roll" of the sound. Some observers feel that this was done, 
but if any improvement was made in decreasing this effect, it was 
quite small. 

The Transient Characteristics of the System. Exceptionally realistic 
quality is not looked for in public address work, particularly under the 
conditions in the Stadium. But from the standpoint of general 
interest, and also to see the effect of a low-pass filter in the amplifiers, 
a frequency run was made at a point on a line at right angles to the 
speaker plane. The distance was again 300 feet, and the result is 
shown in Fig. 8. 

Some Theater Tests. A certain theater that had rather "good" 
acoustics was redecorated. In this process, a good-quality lead paint 
was used on the sound-absorbing material, thereby creating an ex- 
ceptionally reverberant house. In conjunction with the service 
engineers, a study was made of the house to see whether an improve- 
ment could be made by adjusting the distribution pattern of the 
horns, and by a study of the transient response of the system. 

The frequency characteristics of a system in a house whose acoustic 
qualities were considered satisfactory were used for reference. Both 
theaters have identical equipment (W. E. universal bases with 
type 42 amplifiers driving W. E. 555 horn units). 

The data taken are shown in Fig. 9. The two curves with a de- 
ficiency of high frequencies were made at the front sides of the 
house, and the other two along the center. Curves were then run 
on the poor-quality theater, and these are shown in Fig. 10(0). 

Before correction was attempted, a reasonably accurate distribu- 
tion pattern of the house was made; Fig. 11 (a) shows this pattern. 



466 



S. L. REICHES 



Lf. S. M. P. E. 




-5 






^ 







= = 


1 '->. 























































CENT) 


B 




> 








. o S 

S138I33Q 






































^ 






















^^> 








-===: 

RCARCl 


NTER 




., 
















































































































































* 1 I 
FftCOOKM 


1 I 


? i 



FIG. 11. Final curves after correction. 




FIG. 11 (a). Initial distribution curve. 

It will be noted that the greatest level is thrown at the walls near the 
front of the house. 

By using the optimal distribution pattern of this house, the pat- 
tern of Fig. 10(6) was chosen, and the frequency characteristics of 
the system were then corrected as shown in Fig. 11. 

The results obtained were very satisfactory considering the nature 
of the house, and were more than acceptable with a half -filled house. 

Conclusions. From these tests on exponential horns and others 
on the two-way systems, the conclusion has been reached that no 
one type of horn system is universally adaptable. The requirements 
for optimal distribution are the best guide in determining the type 
of speaker system, and once this type is chosen, the most economical 
placing of absorbent material can be found. 



May, 1942] QUARTER- WAVE SPEAKER TESTING 467 

In theaters that are improperly treated, it is often possible to 
improve the sound quality by adjusting the distribution pattern of 
the horns. 

General Conclusions. While all the principles involved in the 
studies presented above are quite generally realized, it is believed 
that the data contained here are quite unusual in that they show 
experimentally the action of speaker systems in the surroundings in 
which they are used. It is further believed that these data could 
not be taken by any method of measurement other than the quarter- 
wave method. The warble tone might be used in a small area of a 
fairly low reverberation period, but in the Cleveland Stadium with 
a reverberation period as long as 10 seconds the warble tone may 
be no better than a steady-state tone. 



SOUND IN MOTION PICTURES* 

NATHAN LEVINSON** 

Summary. A general resume of the technical progress in motion on pictures since 

he introduction of sound. The changes that have occurred in the various technics 

employed are described, including the different kinds of recording, the achievement of 

high-quality records and their reproduction, and the improvements in equipment and 

films. 

PART I 

The sound motion picture has not yet attained such an age that 
many persons will have completely forgotten the thrill which they 
experienced at their first viewing of a talking picture. Yet in that 
brief span of approximately a dozen years since the Jazz Singer took 
the country by storm, the technic of recording sound for motion pic- 
tures, and the equipment and film stocks employed in the process, 
have enjoyed an uninterrupted and almost unbelievable degree o'f 
development. The practice of making duplicate or triplicate sound 
records of a scene to insure a single satisfactory finished record has 
long since been discontinued, and the type of action portrayed on 
the screen is today in no way limited in scope by restrictions im- 
posed by the recording equipment. 

The sound records of the earlier talking pictures were 16-inch disk 
records, similar in general appearance and composition to ordinary 
phonograph records. They were recorded at a rotational speed of 
33 Vs rprn to permit a playing time equal in length to the time re- 
quired for projection of a complete reel of picture. The recording 
channel proper consisted of several condenser microphones and their 
associated amplifiers, a mixer table, booster and main recording 
amplifiers, and a number of bridging amplifiers whose input circuits 
were multipled across a ''bridging bus" formed at the output of the 
main recording amplifier. The output circuit of each bridging am- 
plifier was connected to the cutting head of a wax recording machine 
through a calibrated attenuator. 

* Reprinted from Electronics, XIV (Jan., 1941), p. 17, and XIV (Feb., 1941), 
p. 37. 

** Warner Bros. Pictures, Inc., Burbank, Calif. 

468 



SOUND IN MOTION PICTURES 469 

The cutting heads employed in the production of Vitaphone records 
exhibited a frequency-response characteristic which, for a constant 
input level, produced a record of constant amplitude for all frequencies 
in the interval between 40 and 400 cps, and a record of constant 
velocity in the interval between 400 and 5000 cps. The low-fre- 
quency response of the cutter was reduced to avoid overloading of 
the record by the high-energy, low-frequency components normally 
present in speech and music. The cutter showed a very rapid de- 
crease in response at frequencies above 5000 cps. 

The original records were cut on soft wax blanks whose surface 
had been brought to a high degree of polish by the use of sapphire 
shaving knives. The production of a good record required the use 
of a freshly shaven wax blank whose temperature was held within 
rather narrow limits to prevent smearing or chipping of the wax dur- 
ing recording. The novelty and uncertainty injected in the pro- 
duction of motion pictures by the advent of sound made it necessary 
to provide means for the director of a picture to check the character 
of the sound record immediately upon completion of the shooting of 
each scene. Therefore playback reproducers were provided which 
permitted reproduction of the record cut in the soft wax. Records 
which had been so reproduced were, of course, unsuitable for later 
processing, and for this reason it was necessary to cut two or three 
records of each scene photographed. 

At the completion of shooting and editing of a picture it was 
necessary to combine the individual recordings of each scene which 
appeared in the finished picture in such a manner that the single 
record associated with each reel of the picture would contain just 
that dialog, music, and sound effects necessary for the scenes appear- 
ing in that reel. The process of combining a number of original 
recordings of dialog, music, and sound effects into a single final 
record is known as "dubbing," or "re-recording." The difficulty of 
selecting a few words or sentences from a number of individual disk 
records and combining these in proper order and in exact synchro- 
nism with the action taking place on the screen presented no small 
problem. In fact, the difficulties of this process of selection and com- 
bination were so great that only one of the Hollywood studios, 
Warner Bros., was ever equipped to re-record from disk records on a 
large scale. 

Nor were the troubles encountered in disk recording ended when 
the final records for a picture had been completed. The maintenance 



470 N. LEVINSON [j. s. M. P. E. 

of synchronism between picture and sound in the theater was de- 
pendent upon accurately placing the theater reproducer at the start 
mark on the record and simultaneously placing the picture start 
mark in the picture gate of the projector. If the picture film was 
torn during projection and had to be spliced, it was necessary to 
remove one or more frames of the picture from the reel and conse- 
quently at each splice in a reel, synchronism between picture and 
sound was destroyed by an increasing amount. Since the picture 
and sound record were separate, it was not an unusual occurrence to 
find the record corresponding to one reel of a picture being repro- 
duced with a different reel of that picture. While this may have 
tended to create audience diversion during the screening of a dull 
picture, it helped in no way to maintain the dignity of the theater 
management. Furthermore, after a certain number of playings the 
records exhibited a very pronounced loss of quality and often became 
extremely noisy. All these factors tended to detract materially from 
the technical and entertainment values of a picture and were re- 
sponsible in no small measure for the change from disk recording to 
sound-on-film recording. 

RECORDING ON FILM 

The recording channel employed for producing the early film 
sound-track was practically identical to that employed for producing 
disk records. The signal output of the bridging amplifiers was 
merely delivered to the film-recording machine instead of to the wax 
cutting head. While differing in many details, all film-recording 
machines provide a light-tight housing in which the film is exposed, 
magazines for the unexposed and exposed stock, means for moving 
the film at a uniform speed past a light-beam which exposes the film, 
in accordance with the wave-form to be recorded, the light-source, 
modulator unit, and the optical system. 

The sound-track produced on film varies in width from 76 to 
approximately 100 mils and occupies a position adjacent to one set 
of the film sprocket-holes. All track may be broadly classified as 
being of either the variable-density type or variable-area type and 
each possesses certain advantages and disadvantages not possessed 
by the other. Two methods of producing variable-density track 
were employed during the early period of film recording. In the 
first, typified by the Aeo-light recording at one time extensively 
employed by the Fox Studios, the signal to be recorded modulated 



May, 1942] SOUND IN MOTION PICTURES 471 

the light produced by a gaseous discharge tube and the resultant 
variable-intensity illumination was photographed on the uniformly 
moving film in the recording machine after being passed through a 
very narrow fixed slit. The second and more widely used method 
of producing variable- density track involves modulating a beam of 
constant intensity light by means of a light- valve and photographing 
the illuminated variable-width slit formed by the light- valve ribbons 
on the moving film. This type of record subjects each point on the 
sound-track to an exposure of constant intensity, but of a duration 
determined by the character of the signal being recorded. 

Variable-area track is produced by permitting the light from a con- 
stant-intensity source to strike the mirror of a galvanometer, and 
after reflection therefrom, to pass through a narrow slit of fixed 
width, through a suitable optical system and then upon the sound- 
track being exposed. Oscillations of the galvanometer mirror, which 
are produced by signal currents corresponding to the sound to be 
recorded, cause the light-beam striking the recording slit to illu- 
minate a greater or lesser length of that slit. The sound-track 
produced by this process of recording is essentially an oscillographic 
trace of the signal currents. 

Although the average early sound-on-film records were little, if 
any, better from a quality standpoint than the disk records which 
they replaced, they so facilitated the production, editing, and pro- 
jection of sound pictures that by 1931 practically all sound recording 
was being done on film. Editing the sound record of the finished 
picture was tremendously simplified, since the process of intercutting 
various sound-track sequences presented no greater problems than 
intercutting the corresponding picture sequences. This, of course, 
resulted in enormously simplifying the process of re-recording. It 
was now only necessary to provide reels of properly intercut dialog 
and properly intercut reels of music and sound effects and to re- 
record these in synchronism with the picture to provide a single reel 
of final negative. Film recording provided other advantages, how- 
ever, which were scarcely less valuable than the improvement possible 
in re-recording methods. Unlike the requirements in wax recording 
where many precautions were necessary, the film-recording machine 
could be placed at a considerable angle and be used through a wide 
range of temperature with no change in quality of the finished record. 
The light-valves or galvanometers, once properly adjusted, are com- 
paratively rugged devices and require much less frequent inspection 



472 N. LEVINSON [j. s. M. P. E. 

and maintenance than wax cutting heads. Perhaps the only single 
outstanding advantage of wax over film records lies in the fact that 
the wax record may be immediately played back for checking pur- 
poses, whereas some interval of time must elapse between the record- 
ing process and the time at which completely processed prints from the 
film record are available. 

The introduction of acetate disk recording early in 1934 effectively 
supplemented film recording by providing playback records of much 
greater useful life than soft wax records and having the further ad- 
vantage of possessing more desirable physical properties than wax. 
Continuous improvements in acetate disk coating, as well as im- 
proved designs of cutting heads and reproducers now make it pos- 
sible to produce acetate recordings which are almost equal to high- 
quality film recordings. 

Playback records are no longer employed for checking the record- 
ing of individual scenes of a picture, but find their greatest application 
for reproduction, on the set, of music which has been pre-recorded 
for certain scenes of a picture. The process of pre-recording is 
employed primarily as a means of saving time on the set for such 
scenes of a picture which involve the photography of action which 
must be accurately synchronized with the musical score. For 
example, during the production of elaborate musical numbers in- 
volving complicated dance routines, straightforward production 
technic would demand that the director of the picture divide his 
attention between the action proper, the performance of the or- 
chestra employed, and the degree of synchronism maintained by the 
various groups involved in the complete scene being photographed. 
A flawless performance on the part of the actors could be rendered 
worthless by a slight error on the part of some member of the or- 
chestra, while a perfectly performed musical score might be rendered 
valueless by imperfect synchronism of action on the part of the 
principals appearing in the scene. It is obvious that the difficulty 
involved in securing a completely satisfactory record of such a scene 
is greatly increased by the number of the performing groups. The 
process of pre-recording the musical score for such scenes in a picture 
and reproducing these records on the set while the action is being 
photographed relieves the director of all concern regarding the 
orchestral performance, and permits both the director and the prin- 
cipals involved to concentrate their attention on securing a perfect 
performance. Since the record may be reproduced a number of 



May, 1942] SOUND IN MOTION PICTURES 473 

times with the same results, the scene may be reenacted until a per- 
fect performance is secured. The motors employed for driving the 
playback reproducer and the camera on the set are electrically inter- 
locked, so absolute synchronism between the photographic and sound 
records is assured. 

During the early period of film recording, the quality of the records 
produced was very much inferior to that of present-day sound-track, 
and it is interesting to consider in some detail the numerous improve- 
ments in recording equipment, technic, and materials which have 
made possible the present type of high-fidelity recording. 

The variable-density type of record is essentially a halftone photo- 
graph of the recorded sound-wave. It will be evident, therefore, 
that undistorted reproduction of a variable-density record can be 
obtained only when the entire range of exposure is restricted to the 
straight-line portion of the H&D characteristic of the film employed, 
and when the overall gamma of the print sound-track, as appreciated 
by the phototube in the sound-reproducing mechanism, is equal to 
unity. Although the science of sensitometry was well developed 
long before the advent of the sound picture, little use had been made 
of it in the processing of motion picture films, and the sudden demand 
made upon the laboratory for proper processing of sound-track neces- 
sitated an overnight revision of processing control methods. While 
it was possible for an experienced person to judge the quality of a 
picture negative by inspection with sufficient accuracy for practical 
purposes, this method was wholly inadequate for the determination 
of proper sound-track processing, and had to give way to accurate 
sensitometric control of both negative and print development. The 
introduction of the Eastman type lib sensitometer in 1931 was of 
great value in the study of sound-track processing, since it provided 
a means of accurately and consistently impressing a series of known 
exposures on the film whose characteristics were under investigation. 
So powerful a tool did sensitometric control provide that within a 
few years after its introduction for sound-track purposes, it was al- 
most exclusively employed for the control of both sound-track and 
picture processing in the laboratory. As a result of this step, the 
degree of uniformity and general print quality prevailing throughout 
the motion picture industry today, is almost unbelievably higher 
than that existing in 1930. 

One of the most disturbing characteristics of the earlier film sound 
records, was the high level of film background noise. The average 



474 N. LEVTNSON [J. s. M. P. E. 

level of this noise was determined by the unmodulated track density 
in the case of the variable-density record, and by the width of the 
clear portion of the unmodulated sound-track in the case of the 
variable-area record. The noise level of a typical unmodulated 
sound-track was seldom more than 30 to 35 db below the maximum 
sound level that could be obtained from a fully modulated track. 
As a consequence, those intimate scenes in a picture which required 
the use of relatively low level dialog or background music suffered 
greatly during reproduction. The introduction of sound-track em- 
ploying noise reduction in 1930 extended the volume range of the 
sound record by 10 to 15 db and made possible sound-on-film record- 
ing with a much greater volume range than that which could be 
obtained on disks. Basically, noise reduction on variable-density 
film is secured by making the average transmission of the print 
sound-track proportional to the amplitude of the sound being re- 
corded at any given instant. In variable-area sound-track, to re- 
duce noise, the average width of the clear portion of the track is made 
proportional to the amplitude of the sound being recorded at any 
given instant. 

At approximately the same period during which noise reduction 
was being adopted, the technic of recording had become sufficiently 
standardized so that some thought could be given to the improve- 
ment of frequency characteristics and to microphones, amplifiers, 
and theater speaker equipment. The first of the Western Electric 
moving-coil microphones and of the RCA velocity microphones were 
made available to the industry in 1930. Whereas, it had been 
necessary to mount the microphone amplifier employed with the 
condenser microphone as close to the microphone as practicable, 
the moving-coil and ribbon-type microphones permitted a consider- 
able length of cable between the microphone and the microphone 
amplifier. Microphone boom construction was correspondingly 
simplified and considerably greater ease of following action on the 
set with the microphone resulted. In addition, both the new micro- 
phones exhibited very much better frequency-response characteris- 
tics than did the condenser microphone. 

The first of the so-called wide-range recordings was released in 
1932. These served to indicate not only the added naturalness 
which could be achieved by extending the frequency range, but, and 
what was probably more important, brought to the attention of 
the equipment manufacturers and recording engineers the high de- 



May, 1942] SOUND IN MOTION PICTURES 475 

gree of distortion that existed in the various components of the 
recording and reproducing channels. Whereas the earlier standard 
recordings in many cases exhibited quality which was somewhat 
telephonic in character, the extended-range recordings exhibited an 
unpleasant boominess and excessive sibilance which was extremely 
annoying. Investigations which followed indicated the necessity 
for equalizing the recording channel in such a manner as to decrease 
the low-frequency response on dialog recordings. A portion of this 
equalization has been found necessary to compensate for the dif- 
ference between the dialog level existing at the position of the micro- 
phone during recording and the higher reproduction level existing 
in the theater. Another portion of this equalization, somewhat 
variable in amount, appears necessary to eliminate boominess, or 
low-frequency reverberation, of studio sets. Within the past few 
years, some thought and study have been directed to the determina- 
tion of the character and amount of recording channel equalization 
necessary to compensate for variations in speech effort and cor- 
responding changes in spectral energy distribution of the actors' 
voices during their performances. 

Changes in degree of channel equalization and the insertion of 
low and high-pass niters of various sorts did little more, however, 
than reduce the degree of objectionable distortion existing during 
projection. It, therefore, became necessary to investigate in detail 
the distortion characteristics of each component of the recording 
system as completely as possible. It was soon found that few, if 
any, of the amplifiers employed in the recording channel were nearly 
as free of distortion as had been assumed and a long program of 
amplifier redesign was undertaken. New distortion testing equip- 
ment made it possible to analyze accurately the amount of distor- 
tion caused by the recorder modulator units and by film processing, 
while further studies indicated the need for higher-powered, lower- 
distortion theater amplifier and speaker equipment. 

Amplifier distortion was reduced to an acceptable value by the use 
of transformers having improved frequency-response and impedance 
characteristics, by the use of larger vacuum tubes, by judicious use 
of negative feed-back and in cases where considerable power output 
was required, by the use of carefully balanced push-pull stages. The 
development of heater-type vacuum tubes had progressed to a stage 
which permitted the design of completely a-c operated amplifier 
equipment for the entire recording channel. As a result of these 



476 N. LEVTNSON [j. s. M. P. E. 

improvements in design, the amplifiers employed today have ex- 
tremely low signal distortion at full recording levels and have excel- 
lent frequency-response characteristics. The bridging amplifier em- 
ployed by Warner Bros., for example, has a gain of 11 db at 1000 
cps, with a maximum deviation from this value of but 0.3 db between 
30 and 12,000 cps, and will deliver a power output level of +22 db 
referred to six milliwatts with a distortion of less than 0.5 per cent 
at all frequencies between 60 and 8,000 cps. The combined hum and 
noise level of either of these amplifiers is approximately 85 db 
with respect to six milliwatts. 

The distortion introduced by the recording machine modulator 
unit has been brought to a satisfactory low value by redesign of the 
light- valves and galvanometers and by decreasing the effective width 
of the recording slit image on the film. Further reduction of dis- 
tortion in original recording has been made possible by the use of 
push-pull sound-track. It is interesting to note that one of the first 
patents on push-pull recording was issued in 1911, but no practical 
application was made of this method in motion picture production 
until 1935. 

In Class A push-pull recording two sound records are photographed 
side by side along one edge of the film, one being 180 degrees out of 
phase with the other. Each of the tracks is in itself similar to a 
standard sound-track and by combining the signal resulting from the 
two sound-tracks out of phase, the even order distortion components 
introduced by undesirable film and processing characteristics are 
practically eliminated. 

Class B push-pull recording differs from Class A in that each of the 
individual tracks recorded contains only one-half of the sound-wave 
form. That is to say, all the positive half -cycles of the sound-wave 
are recorded on one of the tracks and all the negative half-cycles 
of the sound-waves are recorded on the other track. During repro- 
duction of Class B sound-track the original wave-form is obtained by 
proper re-combination of the wave-forms appearing on the two sound- 
tracks. 

PART H 

One of the principal requirements which must be satisfied by the 
recording and reproducing mechanisms of a sound motion picture 
system is that of providing absolutely uniform motion of the film 
as it passes the light-beam in the recorder and in scanning beam in 



May, 1942] SOUND IN MOTION PICTURES 477 

the reproducer. Non-uniform motion of the film in either case 
causes frequency modulation of the signal and is particularly ob- 
jectionable during the reproduction of music or relatively long sus- 
tained tones. The frequency modulator, or "flutter," is particularly 
noticeable when caused by rapid acceleration and deceleration of the 
film. The newer recorder and reproducer drive mechanisms have 
been designed to eliminate this type of distortion of providing free- 
running film-loops between the pull-down and take-up sprockets and 
the point of scanning. Critically damped film-driven recording and 
reproducing drums support the film at the point where it passes the 
recorder or reproducer light-beam. Heavy flywheels are provided 
on both the recorder and projector motors to reduce to a .minimum 
any variations in motor speed, and all gears employed for speed 
reduction purposes are carefully ground and fitted to avoid generation 
of speed variations by the gear mechanisms. 

The recording machine designed by RCA utilizes a very simple, but 
most effective, means of securing uniform film motion. The driving 
motor is coupled through gears to a magnet structure which rotates 
coaxially with, but independently of, a heavy flywheel mounted on 
the end of the recorder drum shaft. The magnet structure is driven 
at a slightly higher angular velocity than the normal velocity of the 
recording drum. Eddy currents induced in the rim of the flywheel 
cause the drum shaft to rotate at such angular velocity that the 
peripheral velocity of the drum is just equal to the normal average 
film velocity through the recording machine. In normal operation 
of the recorder, a free-running film loop exists on either side of the 
recording drum, and exposure of any point on the film occurs at a 
point midway along the wrap of the film on the drum. This mecha- 
nism provides an extremely high degree of stabilizing action since 
there is no direct mechanical drive of the film at the actual point of 
exposure. This type of recording machine introduces the equivalent 
of 0.03 to 0.10 per cent frequency modulation as compared with 0.2 
to 0.7 per cent of the older machines. 

The earlier printers used for producing positive sound-tracks by 
contact printing from the negative were found to be a prolific source 
of both frequency and amplitude modulation. A part of the dif- 
ficulty was insufficient contact between the negative and print stocks. 
Slippage of the positive film with respect to the negative at the in- 
stant of print exposure also contributed to the difficulty. The non- 
slip printer design introduced in 1936 provided a means of practically 



478 N. LEVTNSON [j. s. M. P. E. 

removing film slippage during the printing operation. Present-day 
sound printers provide very positive means of maintaining an ex- 
tremely high degree of contact between negative and print stocks and 
utilize an exposing light-beam whose width is of the order of 0.005 
to 0.008 inch. By restricting the length of the sound-track exposed 
at any given instant, the effects of such slippage as may still occur 
are greatly minimized. 

ULTRAVIOLET LIGHT RECORDING 

The extension of the recorded and reproduced frequency range 
requires utilization of the maximum resolving power of the film. 
Research on methods of securing increased negative and print 
definition led to the introduction of a method of recording and print- 
ing which utilizes a narrow band of illumination in the near-visible 
ultraviolet region. When the image of a narrow slit illuminated by 
white light is photographed on film, microscopic inspection of the 
developed image reveals the fact that exposure of the film has taken 
place through the entire thickness of the emulsion. Scattering of the 
light within the emulsion and reflection of light from the film base 
result in the spreading of the image beyond the boundaries of the 
slit. If, however, the slit is illuminated with ultraviolet light and the 
photographic image of the slit microscopically examined, it is found 
that the developed image lies almost wholly on the surface of the 
film emulsion. This occurs because the ultraviolet light is very 
rapidly absorbed in passing through successive layers of the emulsion 
and consequently exposure of the entire emulsion depth is much less 
readily obtained. Correspondingly less scattering of the light within 
the emulsion takes place, and a very much sharper photographic image 
of the slit is obtained. 

The change from white light to ultraviolet recording is accom- 
plished by filtering all the visible radiation from the light-source in 
the recorder, and using a lens system that freely transmits wave- 
lengths as short as 3500 Angstroms. Sufficient ultraviolet radiation 
is obtained from the standard incandescent lamps to secure fully 
exposed sound-track from the ultraviolet portion of the spectrum 
alone. 

The light in the printing machines also is filtered to eliminate the 
visible portion of the spectrum and exposure of the print stock is 
obtained solely by ultraviolet light. 

The application of ultraviolet light to the recording and printing 



May, 1942] SOUND IN MOTION PICTURES 479 

processes has reduced the high-frequency losses to less than 40 per 
cent of the values which were obtained with white light. In ad- 
dition, the lower distortion resulting from improved wave-form on 
the film has resulted in much more pleasing reproduction of high- 
frequency sounds. The reduction of image spread obtained through 
the use of ultraviolet light has also made it possible to increase the 
density of variable-area sound-track prints from an average value of 
1.25 to approximately 1.60. This has resulted in a somewhat lower 
background noise in the theater. 

A very simple method of determining optimal processing condi- 
tions for variable-area sound-track has been devised which involves 
the recording of a 9000-cps signal modulated at a frequency of 400 
cps. All of the 400-cps modulation is removed from the carrier signal 
before the test recording is made, and optimal processing is assured 
when no 400-cps tone is evident during reproduction of the test re- 
cording. Fundamentally, this method of testing enables such a 
choice of print density that the image spread occurring in the print 
cancels that which exists on the negative. A somewhat similar 
method of testing the linearity of the overall processing of variable 
density sound-track has been introduced by ERPI. 

LOUD SPEAKER SYSTEMS 

Most of the early theater speaker systems employed large horns 
equipped with one or more motor units behind the screen. While 
the efficiency of some of these speakers was reasonably high, they 
were deficient in both low and high-frequency reproduction. One of 
the early attempts to overcome the defects of the theater speaker 
employed three speaker units : one for reproduction of the very low 
frequencies, one for reproduction of the middle range of frequencies, 
and one for the reproduction of the extremely high frequencies. 
Suitable dividing networks inserted between the power amplifier 
and speaker terminals provided proper energy distribution to the 
three speakers. Considerable difficulty was experienced with such 
systems in properly phasing and positioning the individual speakers 
so that uniform distribution of energy throughout the theater audi- 
torium could be obtained. Recent developments in speaker design 
have given us the two-way speaker system which employs one or 
more dynamic speakers in suitable baffles for reproduction of all 
frequencies below about 300 cps, and a multicellular horn equipped 
with one or more speaker units for reproduction of all frequencies 



480 N. LEVINSON [j. s. M. P. E. 

above 300 cps. This type of speaker installation has become stand- 
ard in all of the studio review rooms and in all recently equipped 
theaters throughout the country to provide satisfactory reproduc- 
tion of ah* frequencies between 50 and 7000 cps. 

The volume range that may be obtained from a high-quality vari- 
able-area sound-track, such as employed by Warner Bros., is of the 
order of 50 db. For many years it was assumed that naturalness of 
sound in the theater was more or less proportional to the reproduced 
volume range which could be secured from the sound print. It has 
since been found, however, that it is an easy matter to provide too 
great a volume range for satisfactory theater reproduction. The 
general noise level which exists in a theater, caused by normal 
audience movements, heating systems, ventilating systems, and 
operations in the projection booth, determine the minimum sound 
level necessary for a high degree of intelligibility. The type of 
scene portrayed on the screen and general comfort of the theater 
patrons, on the other hand, determine in a general way the maximal 
sound level which may be employed. Studies of a large number of 
theaters have indicated that the difference between the maximum 
level and the minimum level varies between 25 and 35 db. Since 
the volume range existing in the original dialog and music recorded 
for a picture is usually considerably in excess of 40 db, it is evident 
that satisfactory reproduction in a large variety of theaters can only 
be obtained if an arbitrary reduction in volume range is accomplished. 
To this end, electronic volume compressors are installed in each of 
the recording and re-recording channels at the studio, and are nor- 
mally operated so that the original volume range of 50 db is com- 
pressed to a final volume range of the order of 30 db. 

The compressors used in recording are essentially amplifiers whose 
gain is controlled by the instantaneous peak value of the signal 
passing through the amplifier. Gain control is effected by rectifying 
a portion of the signal current and impressing the rectified voltage 
on the control grids of a pair of remote cut-off amplifier tubes in the 
compressor units. The time-constants of the rectifier circuits are 
so chosen that a change in gain of the compressor is accomplished in 
approximately one millisecond. 

FILMS FOR RECORDING 

A resume* of developments in the sound-recording field would be 
incomplete without reference to the advances made in the manufac- 



May, 1942] SOUND IN MOTION PICTURES 481 

ture of film stocks for recording purposes. The early variable- 
density sound-negative records made at Warner Bros. Studio were 
recorded on Eastman type 1301 positive film stock with development 
carried to a gamma of approximately 0.4. This film was originally 
designed for use as a print stock, the development of which would 
be carried to a gamma of 2.0 to 2.4 and was, therefore, somewhat 
low in sensitivity for recording purposes. In September, 1932, the 
Eastman Kodak Company made available type 1359 recording stock 
which had a speed of approximately 2.5 times that of the type 1301 
emulsion. This increase in film speed made it possible to reduce the 
recorder exciter-lamp current by an amount that increased the lamp 
life several hundred per cent, and decreased the variation in negative 
sound-track density which had previously been caused by lamp in- 
stability. 

The 1359 type emulsion was used by Warner Bros, until the intro- 
duction of ultraviolet recording in 1936. At this time the advantages 
of employing variable-area ultraviolet recording appeared sufficiently 
great to justify a complete change in plant recording equipment and 
the RCA variable-area machines were installed. At this time East- 
man made available their type 1357 emulsion which had approxi- 
mately twice the speed of the type 1301 emulsion to ultraviolet light 
and this stock is employed for sound negative at the present time. 

On October, 1937, Eastman type 1360 fine-grain positive film was 
tested as a negative recording stock and found to be somewhat 
superior to the type 1357 film in both high-frequency response and 
background noise. A number of productions were recorded employ- 
ing this stock for the sound negative until it was determined that 
similar improvements could be obtained by utilizing this stock for 
prints employed for re-recording purposes. 

In December, 1939, Eastman announced the replacement of the 
type 1360 emulsion by the type 1361, a film of somewhat lower in- 
herent contrast, and of such spectral sensitivity as to permit handling 
it under positive-type safelights. In all other respects this film is 
similar to the type 1360 emulsion and results in an increased high- 
frequency response of approximately 1.5 db at 9000 cps and a reduc- 
tion in film background noise of approximately 6 db. 

In order to provide negatives from which release prints can be 
made in the various countries, and to provide insurance against the 
possible destruction of the original picture and sound negatives, it is 
customary to prepare duplicate negatives of the picture and sound- 



482 N. LEVINSON 

track negatives by photographic means. The process involves 
making a composite master print from the picture and sound nega- 
tives and by a second printing operation, securing a composite dupli- 
cate negative of the original. Until recently, the composite master 
print was made on Eastman type 1362 lavender stock and a "dupe" 
negative was made from this on Eastman type 1217 panchromatic 
negative stock. Prints made from the duplicate negatives, when 
compared with the original, showed an average increase in film back- 
ground noise of approximately 5 db, a loss in volume of approximately 
2 db, and a reproduction loss of 6 db at 9000 cps. 

In the latter part of 1937, Eastman introduced its fine-grain dupli- 
cating positive stock, type 1365, and a fine-grain duplicating nega- 
tive stock, type 1203. These films have been substituted for the 
lavender positive stock and panchromatic negative stock previously 
employed in making duplicate negatives and prints from this new 
stock show an increase in surface noise of only one db, a loss in 
sound level of one db, and a loss in high-frequency response of only 
one db at 9000 cps as compared to the original. This improvement 
in duplicating stocks represents a remarkable achievement in film 
manufacture and permits the production of prints from duplicate 
negatives that can not be distinguished from prints of the originals. 

It is evident that the exercise of the greatest care and use of the 
latest recording equipment and materials will be of little value unless 
the improvements achieved in recording can be reflected in the qual- 
ity of reproduction obtained in the theater. Warner Bros, has re- 
cently completely re-equipped its entire chain of theaters with the 
latest type of RCA reproducing equipment. This change involved 
the installation of new type sound-heads equipped with rotary 
stabilizers to secure uniform film motion at the point of scanning, 
new amplifiers of greater power handling capacity and lower dis- 
tortion than those previously employed, and two-way loud speaker 
systems capable of reproducing, with a minimum of distortion, the 
entire audio spectrum recorded on the sound-track. While the 
majority of the theater reproducing units are of very rugged con- 
struction, highest quality of reproduction can be obtained only if 
the reproducing equipment is frequently checked and serviced. This 
work is accomplished by a theater engineering service group, and by 
this means it has been found possible to remove likely sources of 
trouble or partially defective equipment before a break-down occurs 
during a performance. 



CURRENT LITERATURE OF INTEREST TO THE MOTION PICTURE 

ENGINEER 

The editors present for convenient reference a list of articles dealing with subjects 
cognate to motion picture engineering published in a number of selected journals. 
Photostatic copies may be obtained from the Library of Congress, Washington, D. C., 
or from the New York Public Library, New York, N. Y. Micro copies of articles 
in magazines that are available may be obtained from the Bibliofilm Service, Depart- 
ment of Agriculture, Washington, D. C., at prevailing rates. 

American Cinematographer 

23 (Feb., 1942), No. 2 
Movies Speed Training of R. C. A. F. Fighting Airmen 

(pp. 54-55, 84) W. STULL 

The Electroplane Camera (pp. 56-57) E. P. HOLDEN, JR. 

Color Television in England (p. 58) J. H. BAIRD 

Motion Pictures in the Army (pp. 69, 84-86) R. B. KONIKOW 

Shooting Technicolor in the Air (pp. 60-61, 86-88) E. G. DYER 

Roy Kellino Films Britain's War Effort (p. 62) W. G. C. Bosco 

Improvising Telephoto Lenses (pp. 72, 93) P. R. NELSON 

Electronic Engineering 

14 (Feb., 1942), No. 168 

Stereoscopic Television (pp. 620-621) J. L. BAIRD 

Frequency Modulation. Pt. IV Frequency Modulation 

Receiver (pp. 628-630, 643) K. R. STRULEY 

Educational Screen 

21 (Feb., 1942), No. 2 
Motion Pictures Not for Theaters, Pt. 34 (pp. 61-63) A. E. KROWS 

Institute of Radio Engineers, Proceedings 

30 (Feb., 1942), No. 2 
Radio Progress During 1941, Pt. 1 (pp. 59-61) 

Electronics; Pt. 4, Frequency Modulation (pp. 65-66); 
Pt. 5, Television (pp. 66-67); Pt. 6, Facsimile (pp. 
67-68); Pt. 9, Radio Wave Propagation (pp. 68-69) 
Factors Governing Performance of Electron Guns in Tele- 
vision Cathode-Ray Tubes (pp. 103-105) R. R. LAW 

Kinematograph Weekly (Ideal Kinema Section) 

299 (Jan. 15, 1942), No. 1813 
Kinema Technique and Equipment. The Outlook for 1942 

(pp. 7, 9) R. H. CRICKS 

Transportable Projectors (pp. 9, 12) R. H. CRICKS 

483 



FIFTY-FIRST SEMI-ANNUAL CONVENTION 

OF THE 

SOCIETY OF MOTION PICTURE ENGINEERS 



HOLLYWOOD-ROOSEVELT HOTEL 

HOLLYWOOD, CALIF. 
MAY 4th-8th, INCLUSIVE 



OFFICERS AND COMMITTEES IN CHARGE 

EMERY HUSE, President 

E. A. WILLIFORD, Past-President 

H. GRIFFIN, Executive Vice-President 

W. C. KUNZMANN, Convention Vice-President 

A. C. DOWNES, Editorial Vice-President 

J. G. FRAYNE, Chairman, Pacific Coast Section 

C. W. HANDLEY, Chairman, Local Arrangements Committee 

S. HARRIS, Chairman, Papers Committee 



R. R. SCOVILLE 
C. R. DAILY 



Pacific Coast Papers Committee 

G. A. CHAMBERS, Chairman 
F. L. EICH 
W. W. LINDSAY. JR. 



S. P. SOLOW 
W. V. WOLFE 



Reception and Local Arrangements 



J. O. AALBERG 
B. B. BROWN 
G. A. CHAMBERS 
W. E. GARITY 

A. M. GUNDELFINGER 

E. H. HANSEN 

J. K. HlLLIARD 

E. M. HONAN 



C. W. HANDLEY, Chairman 

B. KREUZER 

R. G. LlNDERMAN 

C. L. LOOTENS 

R. H. McCULLOUGH 

W. C. MILLER 
G. S. MITCHELL 
K. F. MORGAN 
H. MOYSE 



W. A. MUELLER 
G. F. RACKETT 
H. W. REMERSHIED 
ALSTON RODGERS 
L. L. RYDER 
S. P. SOLOW 
H. G. TASKER 
J. R. WILKINSON 



F. ALBIN 
L. W. CHASE 

484 



Registration and Information 

W. C. KUNZMANN, Chairman 
J. FRANK, JR. 
J. G. FRAYNE 
C. W. HANDLEY 



S. HARRIS 
F. L. HOPPER 



1942 SPRING CONVENTION 



485 



Publicity 

JULIUS HABER, Chairman 
G. R. GIROUX, West Coast Chairman 
L. A. AICHOLTZ S. HARRIS 

J. W. BOYLE S. E. HAWKINS 

J. L. COURCIER 

Luncheon and Banquet Committee 

L. L. RYDER, Chairman 
J. O. AALBERG EMERY HUSE 

J. G. FRAYNE H. T. KALMUS 

C. W. HANDLEY M. S. LESHING 

E. M. HONAN N. LEVINSON 



G. S. MITCHELL 
E. C. RICHARDSON 
R. R. SCOVILLE 



R. H. McCULLOUGH 

W. C. MILLER 

P. MOLE 

H. G. TASKER 



A. C. BLANEY 
D. J. BLOOMBERG 
L. F. BROWN 
J. P. CORCORAN 



Hotel and Transportation 

G. A. CHAMBERS, Chairman 
C. R. DAILY 
C. DUNNING 
W. C. HARCUS 

G. T. LORANCE 



H. R. LUBCKE 

F. O'GRADY 

J. W. STAFFORD 
W. L. THAYER 



J. O. AALBERG 

J. DURST 

G. M. FARLY 

B. FREERICKS 

W. E. GEBHART, JR. 



Convention Projection 

C. L. RUSSELL, Chairman 
L. D. GRIGNON 
J. K. HILLIARD 
A. E. JACKSON 
W. W. LINDSAY, JR. 
R. H. MCCULLOUGH 



S. M. PARISEAU 
H. W. REMERSHIED 
C. R. SAWYER 
G. E. SAWYER 
H. A. STARKE 



Officers and Members of Los Angeles Projectionists Local No. 150 



Ladies' Reception Committee 

MRS. EMERY HUSE and MRS. J. G. FRAYNE, Hostesses 



MRS. G. A. CHAMBERS 
MRS. F. L. EICH 

MRS. A. M. GUNDELFINGER 

MRS. C. W. HANDLEY 
MRS. J. K. HILLIARD 
MRS. E. M. HONAN 
MRS. B. KREUZER 
MRS. N. LEVINSON 
MRS. R. H. MCCULLOUGH 
MRS. G. S. MITCHELL 



Assisted by 



MRS. P. MOLE 
MRS. K. F. MORGAN 
MRS. W. A. MUELLER 
MRS. G. F. RACKETT 
MRS. H. W. REMERSHIED 
MRS. E. C. RICHARDSON 
MRS. L. L. RYDER 
MRS. R. R. SCOVILLE 
MRS. S. P. SOLOW 
MRS. J. R. WILKINSON 



MRS. W. V. WOLFE 



486 1942 SPRING CONVENTION 

Color Print Exhibit Committee 

O. O. CECCARINI, Chairman 

L. E. CLARK C. DUNNING L. D. GRIGNON 

T. B. CUNNINGHAM R. M. EVANS A. M. GUNDELFINGER 

HEADQUARTERS 

The Convention headquarters will be at the Hollywood-Roosevelt Hotel. 
Excellent accommodations have been assured by the hotel management at the 
following per diem rates: 

One person, room and bath $3 . 50 

Two persons, double bed and bath 5.00 

Two persons, twin beds and bath 6.00 

Parlor suite and bath, one person 8 . 00-14 . 00 

Parlor suite and bath, two persons 10. 00-16 . 00 

Room reservation cards were mailed to the membership early in April and 
should be returned to the hotel immediately to be assured of satisfactory ac- 
commodations. 

Registration headquarters will be in the hotel lobby. All members and guests 
attending the Convention will be expected to register and receive their Conven- 
tion badges. The registration fees are used to defray the expenses of the Con- 
vention, and cooperation in this respect is requested. Identification cards will 
be supplied, which will serve as admittance to all scheduled or special sessions, 
studio visits, and trips, and several de luxe motion picture theaters on Hollywood 
Boulevard in the vicinity of the hotel. 

Members planning to attend the Convention should consult their local railroad 
passenger agents regarding train schedules, rates, and stop-over privileges en 
route. For a stop-over at San Francisco the Convention Committee recommends 
the Mark Hopkins Hotel, on "Nob Hill." Accommodations may be arranged 
with Mr. George D. Smith, manager of this hotel. 

An interesting color-print exhibit will be an adjunct to the Convention and will 
be open to the public and delegates during the five days of the Convention. 

The Convention hostesses promise an interesting program of entertainment for 
the visiting ladies. A reception parlor will be provided as then- headquarters at 
the hotel. 

Note: The Pacific Coast Section officers and managers gave serious considera- 
tion to the question of holding the 1942 Spring Convention at Hollywood, and 
have decided to proceed with arrangements for the meeting. The motion picture 
industry plays an essential part from the exhibiting and engineering viewpoint in 
upholding the morale of the general and theater-going public in the present crisis, 
and accordingly the Convention and Local Arrangements Committees are por- 
ceeding with their plans. However, if later deemed advisable in the National 
interest, the Convention will be subject to cancellation thirty days prior to the 
announced Convention dates. 



ABSTRACTS OF PAPERS 

FOR THE 

FIFTY-FIRST SEMI-ANNUAL CONVENTION 

HOLLYWOOD-ROOSEVELT HOTEL 

HOLLYWOOD, CALIF. 

MAY 4-8, 1942 

The Papers Committee submits for the consideration of the membership the follow- 
ing abstracts of papers to be presented at the Spring Convention. It is hoped that the 
publication of these abstracts will encourage attendance at the meeting and facilitate 
discussion. The papers presented at Conventions constitute the bulk of the material 
published in the Journal. The abstracts may therefore be used as convenient reference 
until the papers are published. 

A. C. DOWNES, Editorial Vice-P resident 

S. HARRIS, Chairman, Papers Committee 

R. R. SCOVILLE, Chairman, West Coast Papers Committee 

F. T. BOWDITCH C. R. KEITH W. H. OFFENHAUSER 

F. L. EICH E. W. KELLOGG G. A. CHAMBERS 

R. E. FARNHAM P. J. LARSEN S. P. SOLOW 

J. L. FORREST G. E. MATTHEWS W. V. WOLFE 

Audio-Visual Aids to Naval Training; WILLIAM EXTON, JR., U.S.N.R., Bureau 
of Navigation, Navy Department, Washington, D. C. 

The expansion of the Navy requires expansion in personnel as well as in ma- 
teriel. The expansion in personnel means the provision of literally hundreds of 
thousands of men trained in the operation of complicated equipment. This in- 
cludes radio, various types of ordnance equipment, navigation, the handling of 
ships, tactics and maneuvers, the maintenance and repair of all sorts of equip- 
ment, aeronautical operations of every kind, and every other activity connected 
with naval affairs. 

Naval training is usually conducted by skilled officers, both commissioned and 
non-commissioned. At present, experienced and skilled officers who might nor- 
mally be assigned to training duties are required for operations at sea. At the same 
time, the number of men to be trained has increased enormously, and the number 
of fields in which they must be trained has also increased because of the develop- 
ment of new technics and materiel. Furthermore, new inventions have brought 
about the need for training in fields in which there is virtually no experience and 
concerning which only a few experts have knowledge. 

The Navy believes that the use of audio-visual aids will be of tremendous help 
in this connection. These can not entirely replace the skilled and experienced 
officer or non-commissioned officer, but they have certain very definite advan- 
tages. One of these is the standardization of instruction, so that men trained at 
every activity will have interchangeable skill. Another is the supplementing of 

487 



488 ABSTRACTS OF CONVENTION PAPERS [J. S. M. P. E. 

the instructor who may not have had much pedagogical experience or be an able 
teacher, although he may have ample knowledge of the subject. The third is the 
stimulation of the interest of the trainees. It is possible through visual aids to 
give them a more comprehensive picture of the application of the technic they are 
learning to general naval operations than is normally possible during training. 
Furthermore, since ships are not now available for training purposes as they are 
during peacetime, audio-visual aids make it possible to give some conception of 
the application of the subject in which training is being given to actual operations. 
The armed forces of the United States, guided by such considerations as these, are 
carrying the use of audio-visual aids for training purposes beyond civilian experi- 
ence with these media. It is to be hoped that the experience gained and the 
technics developed will be of value in the future for civilian purposes. 

Motion Pictures: Technology in Art; L. S. BECKER, Warner Bros. Pictures, 
Inc., Burbank, Calif. 

The motion picture and the automobile were born at the turn of the century and 
grew up together. Both have their foundations in science and technology, and 
both have profoundly affected our individual and national lives. Their maturity 
has placed them among the five largest American industries, yet one is fundamen- 
tally an art. An automobile is something concrete, tangible, something real; a 
motion picture is light and shadow, laughter and tears, speech and music. The 
motion picture is an art as well as an industry. The motivating forces of the film 
are drama, comedy, human experience yet it could not exist except for the or- 
ganized efforts of the many craftsmen and technicians that make it an industry. 
Since art and industry are so interwoven, a change in technology affects the art of 
the film, while the demands of the art bring about technical improvements. 

This report illustrates the role that technology plays in the conception of the 
film as an art, and the changes that the demands of the art itself have brought 
about in technic. The cameraman's universal focus, the soundman's reverberation 
chamber, the set designer's cloth ceiling all have their share in telling a story 
realistically and dramatically. Someone's story idea sets this intricate machinery 
in motion, and from the writer, actor, artist, and engineer comes a living entity a 
combination of arts that have been in development since man first learned to re- 
cord his experiences for posterity. 

Motion Picture Laboratory Practices; JAMES R. WILKINSON, Paramount Pic- 
tures, Inc., Hollywood, Calif. 

The function of laboratory service to studio production departments and to the 
release distribution field is discussed. The size and scope of laboratory operations 
are illustrated graphically by an organization chart showing the number of sub- 
departments. These in turn are classified into three major divisions: namely, 
control, processing, and maintenance. Analysis of individual department ac- 
tivity begins with the control division, and emphasis is placed upon the recent 
trend toward the more scientific approach to the problems of processing. 
The discussion continues with the processing division, starting with negative de- 
velopment. The processing method of each successive department is described, 
showing the in-line flow of the work for both studio and release print operations. 



May, 1942] ABSTRACTS OF CONVENTION PAPERS 489 

Problems relating to proper mechanical and electrical maintenance are also dis- 
cussed. 

Re-Recording; L. T. GOLDSMITH, Warner Bros. Pictures, Inc., Burbank, Calif. 

The nature of re-recording as it applies to motion picture production is de- 
scribed in some detail by showing what happens to a typical picture in the re- 
recording department after shooting on the set has been completed and the pic- 
ture has been edited to the satisfaction of the producer. 

Sound is added to those portions of the picture that have been photographed 
silent because of the difficulty or impossibility of recording the corresponding 
sound at the time, as, for example, credit titles, montages, miniatures, stock shots, 
and scenes photographed silent to playbacks of pre-recorded sound. Music that 
has been specially scored and recorded for the picture together with appropriate 
sound effects are added to heighten its dramatic presentation. 

Improvements in dialog quality are made if required by employing electrical 
equalizers; although distortion is often purposely introduced where telephone, 
dictaphone, radio, and similar types of quality must be simulated as required 
by the picture. 

Proper balance of the relative volume of the dialog and the accompanying music 
and sound effects is determined to the satisfaction of the re-recording supervisor. 
All the sounds from as many as a dozen or more different sources are re-recorded 
to a single composite sound-track which afterward is printed with the picture to 
make up the final print to be projected in the theater. 

The organization of the re-recording department is discussed and the duties of 
the various members of the personnel are outlined. Crews are so made up that an 
average of from three to six pictures are in work at the same time. A bibliography 
is included to which the reader can refer for more detailed descriptions of the 
special equipment and processes employed. 

An Analysis of the Complete Sound Recording System Now Used by M-G-M 
Studios; J. K. HILLIARD, Metro-Goldwyn-Mayer Studios, Culver City, Calif. 

This presentation is an illustrated discussion and demonstration of the complete 
variable-density recording system as used by Metro-Goldwyn-Mayer Studios. 

The channel is the result of several years of development and engineering to 
achieve a high standard of quality and signal-to-noise ratio. The complete sys- 
tem will be described, beginning with the microphone, limiter amplifier, R.M.A. 
pre- and postequalization, improved coils with special shielding, cylinder lens, 
noise reduction, improved type headset monitor, fine-grain sensitometry, improved 
super-portable equipment, new recording machines and re-recording machines, 
and disk recording apparatus. Engineering specifications and analysis of the de- 
sign considerations and the functional aspects of the apparatus will be given, 
followed by a demonstration and informal discussion. 

The Development of the Sound-Film, 1927-42; JOHN E. ABBOTT, The Museum 
of Modern Art Film Library, New York, N. Y. 

With the appearance of the part-talking film, The Jazz Singer, in 1927, a new 
motion picture era began. Talking films had been introduced before, but not 
until then had the electrical engineers solved the problem of amplifying as well as 



490 ABSTRACTS OF CONVENTION PAPERS [J. S. M. P. E 

recording and reproducing sound. As often, technical invention preceded creative 
use, and at first the new machines were used clumsily. It almost seemed, momen- 
tarily, that in gaining a voice the movies had lost a soul. Aesthetic pleasure or 
intellectual content are hard to find in the first talkies, although the newsreels 
gained much through the new dimension. The rest was mostly a tinpot substi- 
tute for theater. Yet it is well to recall the character of such primitive talkies as 
Lights of New York (the first all-talkie) in order to understand what the elements 
were that stirred the wonderment and curiosity of the public in 1928, and it is the 
only effective way of appreciating the speed and ingenuity with which this appar- 
ent retrogression in film-making was overcome during the next few years. That it 
has been overcome completely, even today, is perhaps open to question. The 
future is certainly rich in potentialities. 

In gathering together films of the sound era from many countries for its ar- 
chives, the Film Library has necessarily encountered some difficulties. It is not 
generally realized that certain of the early talkies can no longer be projected: 
either the machinery for running them no longer exists or unequal degrees of 
shrinkage in visuals and sound-track (often separate then) preclude synchroniza- 
tion today. Even among some pictures of real importance which survive, only 
poor effects can be obtained. Early sound recording was crude and many factors 
combine to make renewed experience of The Love Parade and Public Enemy, for 
instance, somewhat disconcerting. Audiences following film history at the Film 
Library's showings can not but be sharply aware of the permutations and im- 
provements in sound recording and reproduction or in the dramatic use of sound 
and dialog during the early years of sound pictures. They will recognize that un- 
happy subservience to theatrical form that devitalized the cinema so extensively, 
and rejoice to see it gradually resume the freer, more cinematic technics of the 
silent era. Now many famous film stars go into eclipse, while new favorites 
emerge. Reflecting the world about it, though somewhat belatedly, the cinema 
has already created the gangster film, but sound gives it startling vitality and 
familiarizes the English-speaking world with some strange verbal expressions. 
Now, too, the American film reflects social and political problems, and the trick 
film reappears. The Russian film, continuing to be seen here mostly by special 
groups, hardly overcomes the technical difficulties of sound recording, though its 
directors advance some valuable theories. In France single films flare up again 
and again to remind us of this country's great contribution to cinematography. 
By 1934, the film in Germany was harnessed to the uses of propaganda, with re- 
sults that are important technically and will be valuable historically, though their 
content, highly repugnant to our national taste, precludes their being publicly 
shown. As in previous decades, innovations and advances have come about in 
this country through experiments within the major studios as well as indepen- 
dently in the case of documentary films. Today the motion picture is being used 
by government everywhere as never before and there is no longer any doubt of 
the tremendous importance of this medium to society as a whole. 

The Production of Industrial Motion Pictures; LLOYD THOMPSON, The Calvin 
Company, Kansas City, Mo. 

The production of industrial sound motion pictures is similar to production in 
the major studios. Limited budgets mean that certain short-cuts must be taken 



May, 1942] ABSTRACTS OF CONVENTION PAPERS 491 

but the final screen results must be such that the audience is not aware of the 
limited budget. If satisfactory results are to be obtained, close cooperation is 
required between the director who has his special problems and the technical de- 
partment which also has its special problems. 

The paper lists a number of these problems and also discusses what can be ex- 
pected of industrial producers. 

Procedural and Dimensional Practices for Production of 16-Mm Motion 
Pictures for Television Projection; R. BLACKINTON FULLER AND L. S. RHODES, 
Marsh Cinesound, Inc., New York, N. Y. 

A general report on setting of procedural and dimensional practices for the 
production of 16-mm sound motion pictures for television projection, including 
abstracts from discussion with leading television engineers of the major compa- 
nies for the eventual determination of reproduction standards for equipment and 
methods that are, at the present time, subject to variations that may impair the 
quality or clarity of films projected on the television system. 

The paper shows that in the various steps from the original film to the final 
image on the television receiver, a considerable percentage of the frame area is 
lost by "cropping" in the projector, in the iconoscope, and in the kinescope. Un- 
less this loss is taken into consideration and compensated for in the original plan- 
ning of films for television, loss of image area may seriously impair the effect of the 
motion picture. 

The paper makes specific recommendations based upon the conclusions drawn, 
but does not attempt, in view of present conditions, to fix final aperture standards 
any further than to urge that such standards be set up by the proper group. 
Many of the factors directly concerned in production are considered with a view 
to the ultimate quality to be attained. 

Reference is made to actual experiences and problems met by the authors in 
the preparation of animated cartoons and other films for television broadcasting, 
hoping that their experience may help others to avoid some of the difficulties en- 
countered and thus contribute to the efficiency and effectiveness in the prepara- 
tion of motion pictures for this rapidly growing medium. 

The Practical Aspect of Edge-Numbering 16-Mm Film; H. A. WITT, Wilding 
Picture Productions, Inc., Chicago, 111. 

The use of the edge number and how it is generally applied in the industry, and 
the advantages of edge-numbering at 16 frames as a standard for 16-mm film are 
discussed. 

It has been long-accepted practice to edge-number 16-mm film in relation to 
35-mm frames. Such practice has proved advantageous in complex films, such as 
one constructed of some 16-mm film combined with 35-mm to complete a final 
subject in finished form on 16-mm, still maintaining all the advantages gained 
in the past practice by the use of 35-mm. 

Continuous Replenishment and Chemical Control of Motion Picture Develop- 
ing Solutions; H. L. BAUMBACH, Paramount Pictures, Inc., Hollywood, Calif. 

The chemical reactions that take place in a photographic developer are dis- 
cussed in detail. It is pointed out that, following the determination of a chemical 
formula for producing optimal photographic results, the concentration of every 



492 ABSTRACTS OF CONVENTION PAPERS [J. S. M. P. E. 

important ingredient of the solution may be held constant by continuous replen- 
ishment and chemical control. After a discussion of the theoretical considerations 
involved, details are given for the establishment of picture negative, variable- 
density sound negative, and positive systems in use at the Paramount West 
Coast Laboratory. 

The Application of Potentiometric Methods to Developer Analysis; JOHN G. 
STOTT, Eastman Kodak Company, New York, N. Y. 

Potentiometric titration methods are applied to routine developer analyses in 
order to simplify and speed up the operation and to minimize the "human error" 
arising from judgment of color change end points, etc. A brief theoretical treat- 
ment of potentiometric titrations is included, and new tests for elon, hydroquin- 
one, bromide, and carbonate are outlined. Detailed procedure outlines are in- 
cluded along with a discussion of the problem of pH vs. the alkali content of a de- 
veloper. A glossary showing stepwise procedure operations required to accom- 
plish the analyses has been compiled along with a complete equipment and chemi- 
cal reagent list. The precision of the methods is evaluated by a table showing 
analysis data on carefully mixed known developers. 

The Engineering Aspects of Portable Television Pick-Ups; HARRY R. LUBCKE, 
Don Lee Broadcasting System, Hollywood, Calif. 

The routine of portable television programming may be termed "applied" tele- 
vision engineering. This is hardly more than a byplay of words, but it is intended 
to convey the impression of an engineering technic evolved to put a program across 
regardless of extenuating circumstances. The emphasis is not on engineering, 
but on the program, with engineering as one of the tools used in accomplishing the 
program. 

The essentials of the technic are set forth. Proper preparation requires constant 
servicing of equipment when the latter and the staff are available. A pre-program 
test several hours before program time is essential to consistent performance, and 
allows reasonable time for correcting installation or transportation caused faults. 
A suitable equipment warm-up period precedes the program. Service failures 
during the program are usually unpredictable but must be met by prompt diag- 
nosis and repair. Thorough knowledge of the many circuits, normal and ab- 
normal operational characteristics thereof, and the knack of finding trouble are 
requisites of this aspect. 

Methodical preparation eliminates some of the difficulties. The television en- 
gineering attributes of a program location are tested and recorded prior to the ar- 
rival of equipment. Voltmeter, dummy load, photometer, field glasses, and 
photographic camera comprise the preliminary test equipment. Experiences in 
televising 140 separate portable programs of the Don Lee Television Station, 
W6XAO, Hollywood, are described. 

RCA Audio Chanalyst a New Instrument for the Theater Sound Engineer; 
ADOLPH GOODMAN AND EDWARD STANKO, RCA Manufacturing Co., Camden, 
N.J. 

During the past decade, the technic and equipment of the sound device engi- 
neer have improved tremendously. Progress in this section of the industry has 



May, 1942] ABSTRACTS OF CONVENTION PAPERS 493 

kept pace with other developments, until today the methods and procedures in 
this branch are solidly based upon good engineering practice. 

The growth in this important phase of theater operation has brought with it 
many new and important instruments for more accurate quantitative measure- 
ments of equipment performance. This has led to the demand for a light, compact 
test instrument incorporating the functions of practically all the various instru- 
ments now carried by the theater sound engineer. The requirements for such an 
instrument are met by the RCA Audio Chanalyst. In addition, an entirely new 
service technic known as Audio Signal Tracing has been made available through 
use of this instrument. 

This means that tests and checks can be made on an amplifier or sound system 
while the equipment is operating under normal conditions. In tracing the audio- 
frequency signal, visual and quantitative checks are combined with aural tests. 
The compactness and flexibility of the Audio Chanalyst decrease the time required 
to locate troubles, and the engineer is now provided with new devices to allow him 
to do a more precise and efficient job on a routine service call. 

A One-Ray System for Designing Spherical Condensers; L. T. SACHTLEBEN, 
RCA Manufacturing Co., Indianapolis, Ind. 

A spherical condenser is a simple lens of relatively large aperture. The outer 
portions of such a lens focus the rays much nearer to the lens than do the center 
portions. As a result the lens as a whole fails to produce a sharp image. This de- 
fect is known as spherical aberration. Although no sharp image is produced, an 
image-like region of maximum light concentration does exist. This is known as 
the disk of least confusion. Its diameter may be minimized by shaping the lens 
so as to minimize spherical aberration. It is with this disk of least confusion and 
its required location that the designer of a spherical condenser must deal. 

Without a knowledge of the properties of the disk of least confusion a designer 
might compute rays through a large number of trial lenses until, by an extensive 
and costly trial-and-error process, a condenser having the correct shape for mini- 
mum spherical aberration, with the disk of least confusion at the required loca- 
tion, may be obtained. 

The present paper examines some simple properties of the disk of least confu- 
sion. It shows how, by computing the course of a single ray through the proposed 
lens, a spherical condenser will result having the correct shape for minimizing 
spherical aberration; having also the correct center thickness for its assumed 
diameter and edge thickness; and for which, finally, the location of the disk of 
least confusion is known. The method is applicable to condensers comprising 
more than one lens, and leads to the required design with a minimum number of 
relatively simple trials. 

Developments in Time-Saving Process Projection Equipment; R. W. HEN- 
DERSON, Paramount Pictures, Inc., Hollywood, Calif. 

The projection of a motion picture on a translucent screen for background pur- 
poses has become increasingly important in studio operations during the past ten 
years. Many shots now made through the use of this process would have been 
extremely costly and perhaps impossible if attempted by direct filming of the 
complete action. 



494 ABSTRACTS OF CONVENTION PAPERS [J. S. M. P. E. 

The sharp rise in production costs in the past few years, attributed partly to the 
foreign market situation, demanded that every effort be expended to simplify 
production methods. 

With this in view, Paramount Pictures embarked upon a complete moderniza- 
tion program of the Transparency Department production equipment early in 
1940. New compact projection units, bases for the projectors, rewind tables, 
screen frames, screen-handling equipment, and light-bridges were designed and 
built. This equipment has immeasurably simplified operations as well as im- 
proved quality beyond levels heretofore achieved. 

Specifications and descriptions of this equipment are presented, with emphasis 
upon a comparison of the new with the old. The success of this equipment can be 
attributed largely to standardization of component parts. Complete interchange- 
ability of essential units, coupled with easy access to critical points, has gone far 
toward eliminating lost time and motion in meeting unexpected emergencies. 

Cinematography as Practiced in Hollywood, 1942; JOHN W. BOYLE, Holly- 
wood, Calif., in collaboration with others. 

The purpose of this presentation is to describe current practice in cinematog- 
raphy as followed in the Hollywood studios. Some of the subjects to be covered 
are camera equipment, set lighting, operation of camera crews, exteriors and use of 
booster light, exteriors taken indoors, make-up, diffusion, coated lenses, use of 
light-meters, color contrast of sets, set and production designers, value of hard 
light for exteriors and interiors, stand-ins, air photography, matching stock 
shots, Technicolor and bipack, Kodachrome, and monopack. 

The Focusing View-Finder Problem in Television Cameras; G. L. BEERS, 
RCA Manufacturing Co., Camden, N. J. 

The technical excellence of a television program may frequently depend upon 
the characteristics of the view-finder used in the television camera. Conditions pe- 
culiar to television make it desirable that television camera view-finders be of the 
focusing type. The requirements of an ideal view-finder of this type are dis- 
cussed. During the past ten years a number of view-finder arrangements have 
been investigated in connection with the development of television cameras. 
Several of these are described and their relative merits indicated. 

Some Recent Developments in Record Reproducing Systems: G. L. BEERS 
AND C. M. SINNETT, RCA Manufacturing Co., Camden, N. J. 

Several factors of importance in obtaining satisfactory reproduction of sound 
from lateral-cut phonograph records are considered. An experimental record- 
reproducing system employing the principles of frequency modulation is described 
and data are supplied on the measured and calculated performance characteristics 
of the system. Curves are included showing the vertical force required for satis- 
factory tracking with the experimental frequency modulation pick-up as com- 
pared with other pick-ups of conventional design. 

Frequency Modulation Distortion in Loud Speakers; G. L. BEERS AND H. 
BELAR, RCA Manufacturing Co., Camden, N. J. 

As the frequency response range of a sound-reproducing system is extended the 
necessity for minimizing all forms of distortion is correspondingly increased. The 



May, 1942] ABSTRACTS OF CONVENTION PAPERS 495 

part that the loud speaker can contribute to the overall distortion of a reproducing 
system has been frequently considered. A type of loud speaker distortion that 
has not received general consideration is described. This distortion is a result of 
the Doppler effect and produces frequency modulation in loud speakers reproduc- 
ing complex tones. Equations for this type of distortion are given. Measure- 
ments confirming the calculated distortion in several loud speakers are shown. 
An appendix giving the derivation of the equations is included. 

The Gasparcolor Process; BELA CASPAR, Hollywood, Calif. 

A brief historic review of the photographic multi-layer materials and their ele- 
ments will be given. The principles of the Gasparcolor Process, the first multi- 
layer material which was introduced to the Motion Picture Industry in 1933 will 
be described. The process utilizes a positive printing stock which contains dyes in 
the emulsion layers in the proper densities and color balance; using indifferent 
treating baths for destroying the dye locally and proportionately to the developed 
silver image present in the photographic layers. The process uses fast dyes which 
have good absorption characteristics. 

The processing can be carried out with practically the existing facilities of the 
black and white laboratories, requiring only slight additions. The balancing and 
printing are similar to black and white procedure, the only variables being the 
printing lights, keeping the processing time constant. 

The various steps in the process will be demonstrated and some of. the results 
shown. A brief discussion of the different taking methods suitable for this process 
will also be reviewed. 

A New Sound Motion Picture Reproducing Equipment for Radio City Music 
Hall; J. E. VOLKMANN AND J. S. PESCE, RCA Manufacturing Co., Indianapolis, 
Ind. 

The Music Hall has always maintained a high standard of sound reproduction 
since its opening in 1932. From time to time during this period improvements 
have been made on the original equipment which was the first commercial high 
fidelity reproducer to employ the well known rotary stabilizer sound-head that has 
subsequently become an accepted standard in the industry. 

While this is true, still further improvements were felt desirable to enable more 
forceful presentation of current productions so as to be in keeping with the prog- 
ress made in film recordings. Among these improvements the following are the 
more prominent: 

(1) Greater flexibility between components to facilitate changes in set-ups so 
as to further enhance presentation and also facilitate service, maintenance and 
operation. 

(2} Less distortion and more power output to fully utilize the increased dy- 
namic range of some of the most recent musical recordings. 

(3} Improved distribution of the higher frequencies through the use of a new 
type of horn. 

A review of these requirements indicate that they could be best met and that it 
would be more practical to employ standard components of the latest design. 

This procedure was followed, and early checks on performance show that these 
requirements have been met. 



BACK NUMBERS OF THE TRANSACTIONS AND JOURNALS 

Prior to January, 1930, the Transactions of the Society were published quar- 
terly. A limited number of these Transactions are still available and will be 
sold at the prices listed below. Those who wish to avail themselves of the op- 
portunity of acquiring these back numbers should do so quickly, as the supply 
will soon be exhausted, especially of the earlier numbers. It will be impossible 
to secure them later on as they will not be reprinted. 



1924 



1925 



No. 

19 

20 
21 
22 
23 
24 



Price 
$1.25 
1.25 
1.25 
1.25 
1.25 
1.25 



1926 



1927 



No. 

25 

26 

27 

28 

29 

32 



Price 
$1.25 
1.25 
1.25 
1.25 
1.25 
1.25 



1928 



1929 



No. 
33 

34 
35 
36 
37 

38 



Price 
$2.50 
2.50 
2.50 
2.50 
3.00 
3.00 



Beginning with the January, 1930, issue, the JOURNAL of the Society has been 
issued monthly, in two volumes per year, of six issues each. Back numbers of 
all issues are available at the price of $1.00 each, a complete yearly issue totalling 
$12.00. Single copies of the current issue may be obtained for $1.00 each. 
Orders for back numbers of Transactions and JOURNALS should be placed through 
the General Office of the Society and should be accompanied by check or money- 
order. 



SOCIETY SUPPLIES 

The following are available from the General Office of the Society, at the prices 
noted. Orders should be accompanied by remittances. 

Aims and Accomplishments. An index of the Transactions from October, 
1916, to December, 1929, containing summaries of all articles, and author and 
classified indexes. One dollar each. 

Journal Index. An index of the JOURNAL from January, 1930, to December, 
1935, containing author and classified indexes. One dollar each. 

Motion Picture Standards. Reprints of the American Standards and Recom- 
mended Practices as published in the March. 1941, issue of the JOURNAL; 50 cents 
each. 

Membership Certificates. Engrossed, for framing, containing member's name, 
grade of membership, and date of admission. One dollar each. 

Journal Binders. Black fabrikoid binders, lettered in gold, holding a year's 
issue of the JOURNAL. Two dollars each. Member's name and the volume 
number lettered in gold upon the backbone at an additional charge of fifty cents 
each. 

Test-Films. See advertisement in this issue of the JOURNAL. 



JOURNAL OF THE SOCIETY OF 
MOTION PICTURE ENGINEERS 

VOLUME XXXVIII JUNE, 1942 

CONTENTS 

The Navy's Use of Motion Picture Films for Training 

Purposes W. EXTON, JR. 501 

The Motion Picture Camera in the Army Air Forces 

G. J. NEWHARD 510 

Wartime Conservation in Theater Projection A Con- 
tribution by the Projection Practice Sub-Committee 
of the Theater Engineering Committee 515 

The Defense Program of the Motion Picture Theater 

H. ANDERSON 526 
Technical Progress in the Motion Picture Industry of 

the Soviet Union G. L. IRSKY 532 

The Development of the Sound-Film J. E. ABBOTT 541 
Edwin S. Porter 546 

Concerning Photography as an Art in America 

L. E. VARDEN 549 

Current Literature 554 

Book Review 557 

Highlights of the Hollywood Convention 559 

Program of the Hollywood Convention 562 

Society Announcements 566 

Index of the JOURNAL, XXXVIII (January- June, 

-1942) 

Author Index 570 

Classified Index 573 

(The Society is not responsible for statements of authors.) 



JOURNAL OF THE SOCIETY OF 
MOTION PICTURE ENGINEERS 

SYLVAN HARRIS, EDITOR 
Board of Editors 

ARTHUR C. DOWNES, Chairman 

JOHN I. CRABTREE ALFRED N. GOLDSMITH EDWARD W. KELLOGG 

CLYDE R. KEITH ALAN M. GUNDELFINGER CARLETON R. SAWYER 

ARTHUR C. HARDY 

Officers of the Society 

*President: EMERY HUSE, 

6706 Santa Monica Blvd., Hollywood, Calif. 
* Past-President: E. ALLAN WILLIFORD, 

30 E. 42nd St., New York, N. Y. 
*Executive Vice-President: HERBERT GRIFFIN, 

90 Gold St., New York, N. Y. 
** Engineering Vice-President: DONALD E. HYNDMAN, 

360 Madison Ave., New York. N. Y. 
* Editorial Vice-President: ARTHUR C. DOWNES, 

Box 6087, Cleveland, Ohio. 
** Financial Vice-President: ARTHUR S. DICKINSON, 

28 W. 44th St., New York, N. Y. 
* Convention Vice-President: WILLIAM C. KUNZMANN, 

Box 6087, Cleveland, Ohio. 
* 'Secretary: PAUL J. LARSEN, 

1401 Sheridan St., N. W., Washington, D. C. 
* Treasurer: GEORGE FRIEDL, JR., 

90 Gold St., New York, N. Y. 

Governors 

*MAX C. BATSEL, 501 N. LaSalle St., Indianapolis, Ind. 
**FRANK E. CARLSON, Nela Park, Cleveland, Ohio. 

*JOHN G. FRAYNE, 6601 Roraaine St., Hollywood, Calif. 

*ALFRED N. GOLDSMITH, 580 Fifth Ave., New York, N. Y. 
**EDWARD M. HONAN, 6601 Romaine St., Hollywood, Calif. 

*I. JACOBSEN, 177 N. State St., Chicago, 111. 
**JOHN A. MAURER, 117 E. 24th St., New York, N. Y. 

*LOREN L. RYDER, 5451 Marathon St., Hollywood, Calif. 

* Term expires December 31, 1942. 
** Term expires December 31, 1943. 



Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, included 
in their annual membership dues; single copies, $1.00. A discount on subscription or single 
copies of 15 per cent is allowed to accredited agencies. Order from the Society of Motion 
Picture Engineers. Inc., 20th and Northampton Sts., Easton, Pa., or Hotel Pennsylvania, New 
York.N.Y. 

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 

General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 

Entered as second-class matter January 15, 1930, at the Post Office at Easton, 

Pa., under the Act of March 3, 1879. Copyrighted, 1942, by the Society of Motion 

Picture Engineers, Inc. 



THE NAVY'S USE OF MOTION PICTURE FILMS FOR TRAIN- 
ING PURPOSES* 



WILLIAM EXTON, JR. 1 



Summary. The expansion of the Navy requires expansion in personnel as well 
as in materiel. The expansion in personnel means the provision of literally hundreds 
of thousands of men trained in the operation of complicated equipment. 

Naval training is usually conducted by skilled officers. At present, officers who 
might normally be assigned to training duties are required for operations at sea. 
The number of men to be trained has increased enormously, and the number of fields 
in which they must be trained has also increased because of the development of new 
technics and materiel. New inventions have brought about the need for training in 
fields in which there is virtually no experience and concerning which only a few ex- 
perts have knowledge. 

The Navy believes that the use of audio-visual aids will be of tremendous help in 
this connection. These can not entirely replace the skilled and experienced officer, but 
they have certain very definite advantages. One of these is the standardization of in- 
struction, so that men trained at every activity will have interchangeable skills. An- 
other is the supplementing of the instructor who may not have had much pedagogical 
experience or be an able teacher, although he may have ample knowledge of the subject. 
The third is the stimulation of the interest of the trainees. It is possible through visual 
aids to give them a more comprehensive picture of the application of the technic they are 
learning to general naval operations than is normally possible during training. The 
armed forces of the United States, guided by such considerations as these, are carrying 
the use of audio-visual aids for training purposes beyond civilian experience with 
these media. It is to be hoped that the experience gained and the technics developed 
will be of value in the future for civilian purposes. 



There are four things necessary for victory in war, with respect to 
the armed forces. These four things are (1) numbers of men, (2) 
equipment, (3) training, and (4) spirit or morale. If you have enough 
men, enough equipment, if the men are sufficiently well trained, and 
if they have the right spirit, you will win. It is the Navy's job to 



* Presented at the 1942 Spring Meeting at Hollywood, Calif.; received 
April 6, 1942. 

** Lieutenant (jg), U.S.N.R., Bureau of Navigation, Dept. of the Navy, 
Washington, D. C. 

501 



502 W. EXTON, JR. [J. S. M. p. E. 

build up its personnel to a sufficient number of adequately trained 
men of high morale. These men, manning the two-ocean Navy 
now built and building, can do for the United States what the Navy 
has always done protect our shores by defeating the enemy. 

There is no doubt that motion pictures have an important part to 
play in this program. So far as the Navy is concerned, motion pic- 
tures are of great assistance in the recruiting program, and undoubt- 
edly help to maintain the Navy as a volunteer organization. Some 
of you may perhaps not realize that our Navy has never been manned 
by any other than volunteers. We are proud of that record; and 
the help that motion pictures give in recruiting assists us in main- 
taining today that unblemished record dating back to John Paul 
Jones. 

With respect to equipment : equipment, after all, is the product of 
industry. It is made by men and machines, and is paid for by 
money. We know that motion pictures assist in raising the money 
by stimulating the sales of war bonds and savings stamps. And we 
know that the motion pictures are used in many places to train 
workers. ' Many manufacturers of machinery have caused to be 
produced motion picture training films explaining the use of such 
equipment. Many industrial organizations employ motion pictures 
to help stimulate production. 

The third of the necessities I have mentioned, however, is the 
principal subject of this paper, and that is the contribution of the 
motion picture film to Navy training. I need not remind you that 
modern warfare is technological warfare. It is men and machines 
against men and machines. Modern weapons and modern defensive 
equipment are constantly becoming more complicated. Before the 
days of firearms, men fought hand to hand, and there was little room 
for mechanical complication in the wielding of swords and pikes and 
spears and other such primitive instruments of warfare. The use of 
these weapons could be learned only through long years of practice. 
The man who was experienced with one of these weapons was a pro- 
fessional soldier a mercenary and a few such men could put to 
rout many times their number of unskilled and inexperienced peas- 
antry or burghers who were not accustomed to fighting in this manner. 
With the introduction of gunpowder, there was a tendency to equalize 
the human beings participating in combat, since a prominent noble 
dressed in the most expensive suit of -armor might easily fall victim 
to the bullet of a common soldier. Castles that had stood dozens of 



June, 1942] THE NAVY'S USE OF MOTION PICTURES 503 

sieges over hundreds of years, and were the impregnable homes of 
prominent lords and monarchs, now fell before booming artillery 
pieces. But as wars progressed and cannon opposed cannon, compli- 
cations were introduced to improve the effectiveness of the opposing 
pieces; and thus the handling of these weapons gradually grew be- 
yond the skill of the average untrained man. 

In the early days of our Navy, the range of battle was virtually 
pointblank, and it took no tremendous skill to aim a gun. The 
principal skill in battle was that of the commanding officer, who, by 
maneuvering his ship, brought the other ship within range and ex- 
posed it to the deadliest concentration of fire. Today huge shells 
are sent crashing beyond the horizon, and the swift dive-bomber or 
high-altitude bomber is prepared to drop its deadly missiles within 
a few seconds after being sighted. Under circumstances like these, 
the humblest sailor must if he is to justify his place aboard a modern 
vessel of war be extremely skilled in some useful battle duty. 
Whether he mans a gun or whether in some high position he spots 
and helps to control the fire; whether down in the engine room he 
helps to deliver the essential speed and power; whether he ministers 
to the wounded, handles the communications, or takes care of the 
ammunition supply whatever his battle station, he must perform 
it with the utmost efficiency, since the total result of the efficiencies 
of all the participants in battle is the efficiency of the vessel. Nothing 
less than maximum efficiency can be justified in battle, and such 
efficiency can not be attained when the personnel is not trained to do 
its utmost. 

However, a modern ship of war, and even a modern warplane, 
spends a very small fraction of its life actually in battle. In fact it 
spends a minor fraction of its time even preparing specifically by 
practice for battle. A large part of the time is spent in maintenance, 
and in the pursuits that sustain the life and health of the members of 
the complement of the ship. The members of the gun crew, for 
instance, spend far more time at maintaining the guns and seeing to 
it that the equipment is in perfect condition, than they do in firing it. 
All machinery must be maintained and all the intricate equipment 
for the control and spotting of gunfire, for communication, for detec- 
tion of aircraft and undersea vessels, and for navigation must be 
maintained. 

In times of peace a new recruit was sent to a training station for 
several months of preparation for duty at sea. He was then placed 



504 W. EXXON, JR. [j. a M. P. E. 

aboard a ship, where the petty officers above him as well as his 
commissioned officers would have plenty of time to whip him into 
shape. He would learn from others by doing, and his drag upon the 
efficiency of the vessel was not of great importance. In times of war, 
however, such as the present, a vessel newly commissioned and taking 
aboard a crew that has never worked together before may find itself 
in contact with the enemy in a matter of days. Obviously a wholly 
green and untrained crew can not be allowed to go forth in a war 
vessel, and yet with naval personnel expanded, as it has been, many- 
fold in a very brief length of time, the problem of securing trained 
personnel or of training personnel as secured, is a tremendous one. 
As the number of war vessels in commission increases, the experienced 
skilled personnel is diluted, being scattered among the large number 
of vessels. As more and more ships are required for active duty at 
sea, fewer of them become available for training purposes. 

Thus, though personnel is being expanded beyond all precedent, and 
the need for training was never greater, there is a smaller number of 
skilled personnel available to conduct training, and there is less equip- 
ment available for use in training. Further to complicate the situa- 
tion, in a war like the present, there is an astounding development 
of new technics, of new procedures, new inventions and develop- 
ments, that require the training of thousands and thousands of men 
in the use of instruments about which at first perhaps only a very 
few experts have knowledge. 

The training problems created by these situations are tremendous, 
and in their solution, the visual aids are expected to play and are 
already playing a very important part. I might observe here that 
one of the most important characteristics that is desired in naval 
training is standardization. Men who are graduates of the United 
States Naval Academy have received standardized training, and 
thus an officer aboard one American war vessel can generally predict 
what an officer aboard another Amerian vessel will do under a given 
set of circumstances. This standardization of training is of great 
value. Its value extends down into the field of the skilled enlisted 
men, since the interchangeability of men is of importance to the 
efficiency of the fleet; and a man who has learned to do a thing a 
certain way on one ship and is expected to do it another way upon 
another ship will not be giving to the Navy the fullest benefit of his 
experience. If men are taught by other men, there is always the 
tendency away from standardization, since each individual has his 



June, 1942] THE NAVY'S USE OF MOTION PICTURES 505 

own idea of what should be stressed and how things should be done. 
However persistently the Navy itself may foster standardization, 
there is a trend away from standards where teaching is done by 
individuals. 

Audio-visual aids, however, help to standardize. Since they can 
be used throughout the naval service and since they will appear 
identically to all who see them, they have the most helpful effect in 
standardizing training. Furthermore and this is extremely im- 
portant audio-visual aids can standardize training on a high level 
rather than on an average level. It can not be denied that there are 
good teachers and bad teachers. Some men are skillful and others 
not quite so skillful in training men. If audio- visual aids to training 
are prepared by the best available experts, and are properly designed 
to have maximum value for training purposes, then training through 
them can be standardized on a very high level. It is hoped that as 
the use of audio- visual aids in the Navy develops, this will be in- 
creasingly true; and the Bureau of Navigation is extremely desirous 
of setting and maintaining a high standard for the visual aids pro- 
duced for use in naval training. 

I should like to stress the point that visual aids to training are not 
entertainment. They are not intended to be entertainment, and 
they should not be considered as in any way related to entertainment. 
The mere fact that an audio-visual aid may be a motion picture 
should not cause it to be confused with motion pictures produced for 
entertainment purposes. The motion picture is a use of a photo- 
graphic technic which can serve many purposes. The fact that it 
has served the purpose of entertainment so greatly should not in- 
fluence, or better, should not impair its use for training purposes. 

There is a tendency on the part of some of those who attempt to 
produce films for training purposes to make these films approximate 
films for entertainment. They introduce the films with music, and 
have music arising many times during the course of the film; and 
introduce elements of incidental comedy, and in other ways try to 
make the film, as they would say, palatable. 

This is a gross abuse of the principle of the training film. A train- 
ing film should be regarded as a text-book. It should be easy to 
understand, it should be clear, it should be simple, it should introduce 
no unnecessary complications ; but there is no obligation on the part 
of a text-book to be amusing or ingratiating. Furthermore, the 
proper use of a training film will usually involve its being repeated. 



506 W. EXTON, JR. [J. S. M. P. E. 

Most of the training films that I have seen can best be used by show- 
ing them a number of times, perhaps giving the men opportunity to 
ask questions or to be lectured to between the showings. A fairly 
complicated film, which gives the men only a rough idea of the subject 
the first time it is shown, may be very simple and easy to understand 
after it has been shown several times, and all the questions asked 
have been answered, and the subject has been explained. A film 
that is well conceived and executed will be just as interesting the 
fourth or fifth time it is shown as the first time, and a new instruc- 
tional benefit will accrue from each showing; but a film that is made 
to be entertaining is very likely to look extremely silly the second 
time it is shown, and thus defeat its purpose. 

Those of you whose experiences have been primarily in the field of 
making films for entertainment should bear this very much in mind 
if you venture into the field of training film. No doubt many of you 
will, as this medium achieves greater and greater use in civilian as 
well as in naval and military training activities. 

The Bureau of Navigation has a Training Division. This Division 
has cognizance over all training of naval personnel, officers, and men. 
We conduct an extraordinarily large number and diversity of training 
activities. We run the Training Stations to which the newest re- 
cruits are brought for their first experience with the Navy after the 
Recruiting Station. We run the various schools where enlisted men 
are taught the trades that make them experts in a very large number 
of fields. We train the Navy's officers at the U. S. Naval Academy 
at Annapolis and at the Naval Reserve Officers Training Corps Units 
in the many universities, and at the Reserve Midshipmen Schools. 
We conduct the Postgraduate School, and the Naval War College, 
at which the higher officers receive advanced training. We conduct 
schools that teach about Diesel engines, underwater sound, radio 
material, torpedoes, submarines, gunnery, and all the other specialties 
that modern naval warfare involves. 

It is a part of the task of the Bureau of Navigation to find for these 
many activities the audio-visual aids that will help the officers 
commanding these schools, and their staffs, to do their jobs as well 
as possible. The audio-visual aids sent to these many activities 
come from, a great diversity of sources. Officially, the agency for 
the procurement of photography in the United States Navy is vested 
in the Bureau of Aeronautics; and if the Bureau of Navigation 
wishes to procure a film to send to a naval activity, we request the 



June, 1942] THE NAVY'S USE OF MOTION PICTURES 507 

Bureau of Aeronautics to procure that film. If it is a question of 
procuring that film by having it produced, then we designate an 
officer to represent the Bureau of Navigation as our expert in the pro- 
duction. This officer then controls the production, working with the 
commercial producer contracted with through the Bureau of Aero- 
nautics. In a few instances, the films may be made entirely by 
naval personnel, since many of our enlisted men are being trained as 
cameramen, and are being used for various purposes not appropriate 
to civilians. 

Individual naval officers at various places have manifested consid- 
erable originality and initiative in developing visual aids for specific 
purposes. I have seen, for instance, a combination of motion pic- 
ture, film-strip, and disk recording that was used to teach some of the 
principles of the detection of submarines by underwater sound. I 
have also seen an excellent sound film-strip on Man Overboard Drill, 
made by an officer who teaches seamanship at one of our training 
schools for young officers. There are other examples of this initiative 
in the development and use of audio- visual aids. The Aviation 
Service Schools train the many thousands of mechanics needed to 
serve our naval aviation, and the officers conducting these schools 
have developed an elaborate collection of visual aids, motion picture 
films, and film-strips covering almost every subject taught in these 
schools. They have even developed a series of visual aids for teaching 
the teachers how to teach. This is useful, since many of the men 
giving instruction in the Aviation Service Schools have not had pre- 
vious teaching experience, even though they do know their subjects 
thoroughly. 

All the films produced by the Army Signal Corps or by the Research 
Council of the Motion Picture Academy of Arts and Sciences for the 
War Department are reviewed by the Navy, and, if they have an 
application, are utilized by the Navy. The Navy also uses films 
produced by other agencies of the United States Government, such 
as the United States Office of Education. This organization recently 
produced and is still producing a number of films on machine shop 
practices, showing how to use various machine tools. These have 
considerable value, and are used in many of our schools. 

There is a number of commercial producers of training films, some 
of whom have produced films on their own initiative or for the use 
and distribution of large industrial concerns. Many of these films 
have been evaluated, and some of them are useful to the Navy. I 



508 W. EXTON, JR. [J. s. M. P. E. 

may add here that where these films are not confidential in nature 
and may have applications for public use, they are to be made avail- 
able for public distribution. 

There is another source of training film material, which has not 
been very deeply tapped as yet, and that is the enormous amount of 
films that have already been made in one connection or another, and 
which, by cutting and splicing and editing, can be made into useful 
training material. There are a number of films made by the British 
government and for the instruction of the personnel of their armed 
forces, and some of these have considerable value. All of these are 
evaluated for naval purposes and the Navy is now using audio-visual 
aids from all the sources mentioned. 

I can not close this paper without referring to the contribution of 
the motion picture to the fourth of the four essentials to victory that 
I mentioned spirit, or morale. The United States Navy has never 
had to worry about the morale of its men, but that does not mean 
that the motion picture can not contribute greatly to the maintenance 
of American naval morale. The motion picture show aboard ship is 
a regular institution, and it is never omitted if it can be helped. This, 
of course, is for entertainment purposes, but the motion picture now 
also serves the purpose of giving to the American bluejacket some 
idea of our allies the United Nations fighting side by side with us. 
The motion picture also helps to remind the American bluejacket of 
the country that is behind him, of the activities of its civilians, and 
of the other armed forces, and of the outward manifestations of the 
basic principles of democracy that are fundamentally involved in 
this world conflict. 

So you can see from this little talk that the motion picture is im- 
portant to the Navy. You men who are important to the motion 
picture should remember this and help to make that importance 
greater, more significant, and ever more contributory toward our 
eventual victory. 

In conclusion, I should like to leave this thought. One of the most 
important aspects of civilization is the transmission of ideas. The 
motion picture is, as you know well, a potent factor in this field. 
Its use for entertainment has been developed to a very high degree; 
its use in training and in education is only beginning. And yet the 
application of the motion picture to training and to education may 
in the end be far more important than any of us imagines at present. 
Even now it may be of vital importance in building the efficiency of 



June, 1942] THE NAVY'S USE OF MOTION PICTURES 509 

our armed forces today ; and it can surely play a great role in develop- 
ing the minds and spirits of young and old in the future. 

In the past the motion picture industry has not lent its best brains, 
its greatest talent, its most creative potentialities to the making of 
films for purposes of training or education. This use of the motion 
picture film, however, is fully worthy of the closest attention of the 
ablest technicians and the greatest brains this industry can boast. 
The rewards may not be measured in dollars to the same extent that 
the production of films for entertainment may be measured in dollars ; 
but the nature and quality of the contribution made, and the perma- 
nent value of that contribution will bring satisfaction that can not 
be compared with that of the ephemeral success represented by a 
"hit." 



THE MOTION PICTURE CAMERA IN THE ARMY AIR 

FORCES* 



GUY J. NEWHARD' 



Summary. A brief account of the various uses of the motion picture camera in the 
Army Air Forces. A number of examples of such uses are given and it is pointed out 
that new demands coming in regularly make it reasonable to believe that the only limi- 
tation of the use of motion pictures in the total war effort will be the number of cameras 
and cameramen available. 



Despite a complete lack of romance, glamor, and box-office names, 
regardless of highly technical themes, no one goes to sleep at an 
official screening in the Motion Picture Branch at Wright Field. 
Wright Field is the research and development center of our Army 
Air Forces. Aerial warfare touches the lives and thoughts of every 
person in the world; no topic holds such universal interest unless it 
be the day-to-day progress of the war itself. 

Civilians rarely see the technical subjects we film. We make 
movies of tests when the action involved is too fast, too complicated, 
or too remote for accurate observation with the human eye. The 
motion picture camera is the one instrument that can capture the 
full time-sequence of some types of experiments. The same thought 
must have occurred to the men who made The Great Train Robbery. 

So, when the pictures of a critical test flash upon our screen, the 
engineers who usually make up the audience are all attention. Pos- 
sibly they pay closer attention in case of a failure, because the film 
will tell them when, and where, and why, and how the failure hap- 
pened. But never do any of the aeronautical specialists aircraft 
engineers, structural and designing experts, engine, propeller, arma- 
ment, equipment engineers fail to give full concentration. They 
make the best audience in the world. 

* Presented at the 1942 Spring Meeting at Hollywood, Calif. ; received April 
13, 1942. 

** Major, Chief of Motion Picture Branch, Materiel Division, U. S. Army 
Air Forces, Wright Field, Dayton, Ohio. 
510 



MOTION PICTURES IN AIR FORCES 511 

The Motion Picture Branch makes all aerial pictures for the Army 
Air Forces. It also does much work in laboratories and on location at 
various airdromes, filming unrevealed projects with an unrevealed 
number of camera crews. Its personnel includes a number of men 
who have worked in Hollywood, and its development laboratory has 
much of the latest equipment employed in the industry's shops. 

Our camera equipment is most complete. Akeley and Eyemo 
cameras are used in the air ; Mitchells and Bell & Howells for ground 
work. One to eighteen-inch focal-length lenses are regular acces- 
sories. A most complete line of all 16-mm equipment is also used. 
Three large Fonda developing machines are used, plus rack and tank. 
Before this year is finished, the laboratory will have processed mofe 
than 20,000,000 feet of film, a decided increase over last year's 
1,000,000 feet. Depue and Bell & Howell printers are used through- 
out. Our own optical printer for special effects and four channels of 
RCA plus two portable units complete our sound department. 

It is no secret that the armed forces of every country are making 
more technical films than ever before, just as a larger number of 
morale and propaganda subjects are being made for public showing. 

To cite a few instances illustrating how motion pictures aptly aid 
and expedite military aircraft development: 

Not many weeks ago an experimental dive bomber crashed during 
the final tests. The factory test-pilot was killed. Ground observers 
who watched the new ship drop like a plummet from 20,000 feet in a 
terminal velocity dive power dive advanced various causes which 
were not in agreement. 

From the wreckage film was recovered intact from a camera and 
this revealed not only the initial cause but clearly pictured the prog- 
ress of each successive failure, step by step, starting with the pull-out 
until the crippled plane shattered itself on the ground a few moments 
later. This film is good "red meat" for structural and design en- 
gineers at Wright Field, and in the plants where very high perform- 
ance airplanes are built. 

Much like one would sight in a new gun, new bombardment air- 
planes are put through bombing practice, using the various sizes of 
bombs. The purpose is to observe the trajectories of light bombs, 
which will be a little different from those of medium bombs, which in 
turn fall unlike heavy bombs. 

Excessive turbulence within the bomb bay may cause the bomb to 
wobble just after it is released from the rack before its fins steer it into 



512 G. J. NEWHARD (J. & M. t>. E. 

steady downward flight. On such practice missions the motion pic- 
ture camera records, for the study of armament engineers, exactly 
what happens. The camera follows the bombs clear down to the 
target, and furnishes data from which are calculated the number and 
size of bombs most effective against various targets. 

In windrtunnel tests cameras are used to show how an airfoil twists 
or bobs and flaps at different air speeds under varying loads. For 
example, a new design of wing is weighted near the root to simulate 
different engines in various positions, while farther out on the wing 
weights are added to represent the machine gun and ammunition 
load. From a study of the patterns and movements at various 
speeds, engineers can calculate the most efficient engine position, 
armament location, and the maximum load the wing can support with 
safety. Streamlining of airfoils, nacelles, airscoops, and other units 
has been improved after motion pictures of smoky airflows around 
the parts suggest a way to improve the design. 

Before any new warplane goes into production it is customary for 
the manufacturer to send what is known as the skeleton article to 
Wright Field for structural tests. In such tests, bags filled with lead 
shot are piled upon all parts of the airplane subject to stress in flight, 
to prove whether the airplane can withstand every maneuver it will 
be put through in military use. To record the failures that normally 
come some time after the design load is exceeded, only a high-speed 
motion picture camera taking 2200 exposures per second can detect 
where the first failure occurs and follow the path of successive failures 
as the structure collapses. 

The camera is used in another type of test in the Structures Lab- 
oratory. This is known as the drop test, in which a fully loaded air- 
plane is dropped vertically from various heights to determine the 
strength of the landing gear. The film record is obtained by mount- 
ing small lights at points from which can be measured the deflection 
of the tires and oleo struts under impact. 

Comparative tests of experimental parachutes with service chutes 
are made by dropping the chutes simultaneously from an airplane at 
various altitudes and speeds. Each chute carries a dummy of known 
weight. On the ground below the cameraman obtains a record from 
which equipment engineers can tell how long each chute requires to 
open, how the experimental chute compares in stability and oscilla- 
tion, and how long it takes to lower a given load to the ground from a 
known height. The dial of a stop-watch is shown on each frame. If 



June, 1942] MOTION PICTURES IN AlR FORCES 513 

the experimental chute rips when dropped with heavy load at high 
speed, the camera record makes the best kind of testimony to show 
the hopeful manufacturer why he can not be given a contract. 

In a similar way, with the exception that cameramen are used in 
the air as well as on ground, thousands of feet of film are shot during 
the development of equipment for parachute troops. Every detail 
of different types of chutes, static cords, jumping technics, cargo 
parachutes, and methods of laying down large forces of parachute 
troops around an objective was filmed to detect and eliminate un- 
necessary hazards due to malfunctioning of equipment. 

An assignment from the Chief of the Air Forces to cover the large 
scale maneuvers down South put our cameramen overhead in photo- 
graphic airplanes to bring back a panoramic aerial view of the activ- 
ities of both the Red and the Blue Armies. This film shows the 
movement and counter-actions of tanks, trucks, troops, and air forces 
en masse. And it discloses whether any of the forces were exposed 
unnecessarily to attack from the air, whether aerial attacks were 
executed with full effectiveness, and how the master strategy of each 
side worked out. 

One instance of the camera's aid to aviation medicine may be 
described. This is the set-up that shows close-ups of pursuit pilots 
during "blackouts" brought on by tight turns at high speeds, pull-outs 
after dives and other types of extreme acceleration encountered in 
aerial combat. Besides the facial contortions and body sagging, 
instruments in the installation show the airspeed, altitude, duration, 
and force as some of the details required by the medical officers. 

This film does not have the audience-suspense of some of the 
special-effects shots that were made for Test Pilot unless one knows 
the conditions under which the official films were made and their pur- 
pose. 

When chemical warfare experts modify equipment for use in mili- 
tary aircraft, motion pictures are very helpful in finding out the re- 
sult. For example, here is a job of installing smoke screen equipment 
in a new and faster type of attack airplane. Assume that the mission 
of this plane is to streak across the edge of an enemy airdrome a few 
seconds before the attacking planes come in to bomb and strafe 
hangars, gas trucks, grounded planes, and defensive gun emplace- 
ments. It is highly important that an adequate curtain of smoke be 
placed to protect the following planes from defensive fire. So the 
cameraman loads into a rear cockpit for practice runs, and the pilot 



514 G. J. NEWHARD 

lays down smoke curtains around imaginary objectives. In the pro- 
jection room the chemical warfare officer can readily see whether the 
smoke screen is high and wide enough for its purpose, and whether it 
has open gaps. He can then prescribe the operating instructions for 
the equipment under the various wind conditions encountered in 
actual combat, and determine the proper capacity. 

One of the Air Corps items tested by machine gun fire is the bullet- 
proof fuel tank. Everything that happens to the bullets and the 
tanks must be carefully scrutinized by the engineers. Each factor- 
first contact of the slug on the near side, the hole that is drilled, the 
change in the shape of the slug, its tumbling motion through the tank, 
and the jagged hole torn in the far side of the tank as the slug emerges, 
are all elements that must be considered. A complete test alternates 
single shots ja.nd bursts. The rate of fire of the machine guns is 
possibly 700 shots a minute. So, while still pictures serve to show 
what the tank looks like at the completion of the test, it is necessary 
to use high-speed motion picture cameras to obtain complete data. 

As a matter of fact, aircraft guns themselves have on occasion been 
improved through pictorial evidence secured with a motion picture 
camera. For example, a certain gun had a slower rate of fire in its 
early tests than was expected. A series of high-speed motion pic- 
tures was made of the gun action, and close analysis showed that one 
key part was binding. When this was given a little attention the 
sluggishness disappeared. 

Motion pictures have been keeping the historical record of the 
progress of aviation for over three decades. Perhaps the oldest 
print in the storage vaults is one made when the Wright Brothers gave 
a demonstration flight for the Army at Fort Meyers in 1908. Many 
of the aviation "firsts" since filmed were included in Cavalcade of 
Aviation recently released by Universal. 

I have mentioned only a few of the ways in which the motion pic- 
ture camera is serving the Army Air Forces. New demands coming in 
each month make it reasonable to believe that the only limitation to 
the use of motion pictures in the total war effort will be the number of 
cameras and cameramen available. 



WARTIME CONSERVATION IN THEATER PROJECTION* 

A CONTRIBUTION BY THE PROJECTION PRACTICE SUB-COMMITTEE 

OF THE 

THEATER ENGINEERING COMMITTEE 

In collaboration with the War Activities Committee of the Motion 
Picture Industry, Richard Walsh, president of the IATSE, recently 
announced a ten-point program designed to conserve vital materials 
needed for military purposes; to salvage such materials; and, by 
reducing waste to a minimum, enable the motion picture theaters 
to carry on during the present emergency. 

In a message accompanying the printed program distributed to 
the theaters of the country, Mr. Walsh said, "Our country is at war. 
Here's how you can help. Every type of material is required in 
America's war effort. Many materials which you handle every day 
are scarce. Spare parts are hard to get. Your theaters may have 
to close unless the equipment that you handle is cared for and con- 
served. It is vitally important to maintain your projection, sound, 
and stage equipment in good operating condition. Only in this way 
can your theater be kept open to do its vital job of maintaining 
morale. Conserve, Salvage, Eliminate Waste." 

The 10-Point Program is as follows: 

(1) Keep your projection rooms and equipment clean. Dirt causes wear and 
tear. 

(2} Lubricate properly all equipment. Follow the manufacturer's instruc- 
tions. 

(3) Make only necessary replacements to conserve spare parts. 

(4) Burn carbons at minimum current specified by manufacturer. Use car- 
bon savers where available. 

(5) Clean lenses of optical systems with soft tissue and protect condensers 
and reflectors. 

(6} Service regularly all electric current distribution points, such as: motors, 
generators, bus bars, fuses, switches, resistors, and condensers. 

* Presented at the 1942 Spring Meeting at Hollywood, Calif., and at the Meet- 
ing of the Atlantic Coast Section at New York, May 21, 1942: 

515 



516 WARTIME CONSERVATION IN PROJECTION [j. s. M. p. E. 

(7) Allow sufficient warming-up period for all vacuum tubes. Burn tubes at 
specified ratings of equipment manufacturers. 

(8) Inspect, thread, and rewind film very carefully. Keep it clean. 

(5) Handle reels and film containers with care; these can not be replaced. 
(10} Do Not Throw Anything Away. 

Keep all worn out parts and metal coated carbon stubs; collect copper and 
other carbon drippings. Keep all burned out or broken vacuum tubes and 
incandescent lamps. You will receive instructions as to the proper disposal 
of this salvaged material. 

The Projection Practice Sub-Committee of the SMPE Theater 
Engineering Committee is wholeheartedly in agreement with the ten 
points and their aims and purposes. However, the Committee feels 
that the value of the ten points would be greatly enhanced if the pro- 
jectionists of the country were informed more in detail of the ways 
and means of accomplishing the ten points. There is much beneath 
the surface in each of the points, and to bring out clearly all the de- 
tails underlying the wordings of the points, the Projection Practice 
Sub-Committee has prepared the following elaboration of the ten- 
point program. Each of the following sections was prepared by a 
member of the Projection Practice Sub-Committee and the separate 
contributions were then correlated to present a clear picture of the 
great amount of detail that underlies each of the ten points : 

(1) Keep Projection Rooms and Equipment Clean. Dirt Causes 
Wear and Tear. Dirt has been the cause of serious film fires in pre- 
venting the proper operation of the automatic fire-shutter or in 
clogging the fire- valve rollers. It makes them susceptible to wear and 
renders them useless for the purpose intended. 

Dirt may cause the stoppage of sound reproduction by accumulat- 
ing on the various movable contacts or on the vacuum-tube contacts 
in the sound equipment. 

It may cause losses in screen illumination, when deposited on the 
projection arc reflector or condensers, and has resulted in the rapid 
deterioration of carbon contacts with communicated damage to the 
adjacent parts of the lamp mechanism. 

Dirt on the gear-teeth and shafts of the projector, combining with 
the lubricating oil, acts like a grinding compound, causing excessive 
wear and shortening the effective life of the gears and bearings. 

On fuse-clips it causes high-resistance contacts and the generation 
of heat, which may sometimes cause the fuse to blow. 

Make sure that the lamp house and all parts are kept thoroughly 



June, 1942] WARTIME CONSERVATION IN PROJECTION 517 

clean both inside and outside. The carbon ash, drippings, etc., should 
be removed regularly once a day, especially from the shafts, bushings, 
and gears of the arc control operating parts. 

The arc exhaust dampers and ducts should frequently be cleaned 
thoroughly of carbon ash, dust, etc. Any blockage, no matter how 
small, will affect the proper burning of the carbons, cause pitting 
of the mirrors, and produce a gradual accumulation of ash within the 
lamp house. If there is a filter in the air-supply system, make sure it 
is in efficient working order. Care should be taken to prevent dust 
and dirt from blowing into the projection room through any windows 
if left open. 

To get the most out of motor-generators, they should be kept clean, 
and all dirt should be removed before sparking becomes disastrous. 
Increased brush life as well as increased commutator life will be the 
direct result. Dirt on commutators causes arcing and pitting, short- 
ening their life and increasing maintenance costs. The contacting 
surface of each commutator brush should be periodically examined 
so that commutator and bearing wear is held to a minimum. If the 
generator is on a concrete floor, care should be taken in sweeping, so 
that abrasive dust from the concrete will not get into the bearings. 
The exhibitor who is interested in keeping his projection maintenance 
costs low should extend to the projection room the same services used 
in cleaning the auditorium and other parts of the theater. The pro- 
jection room floor, walls, and ceiling should be of such materials that 
they will not "dust off." If the floor is of exposed cement, it should 
be kept well painted with "dust-proof" or "sealer" paint, and should 
be mopped frequently. A supply of lintless cloths for cleaning should 
be made available, as well as other cleaning facilities such as carbon 
tetrachloride, brooms and dust pan, metal waste can, and the like. 
In fact, these should be standard equipment of the projection room. 

A stiff-bristled tooth-brush is useful for keeping the sprockets 
and idler rollers clean. The space between the fire-valve rollers and 
the castings in which they are mounted can easily be cleaned by in- 
serting a narrow strip of film and drawing it back and forth to dislodge 
the dirt. 

(2) Lubricate Equipment Properly. Follow the manufacturer's 
instructions, and use only the grade of oil recommended by the manu- 
facturer. The importance of lubrication of projection equipment 
can not be overemphasized. Now that metals and oil have become 
important in our country's war program, we must regard the lubri- 



518 WARTIME CONSERVATION IN PROJECTION [j. s. M. P E. 

cation problem from the conservation viewpoint as well as the oper- 
ating. 

Projection equipment lubrication carried out properly and under 
manufacturers' instructions will lead to trouble-free operation. 

The use of the proper types of oils and greases and their proper 
application will give longer life to the equipment and keep the stand- 
ards of projection on a high plane. 

The following rules should be strictly adhered to : 

(1) Do not lubricate the mechanism while it is in motion. Doing so is 
hazardous both to the mechanism and to the projectionist. 

(2) Do not over-lubricate. Excessive lubrication is costly and wasteful. 
It also impairs the quality of the sound and the picture. Only small oil cans 
that dispense small quantities of oil at a time should be used. 

(3} Cleanliness in conjunction with lubrication is an important matter, 
since excess oil deposits promote the collection of dirt, dust, and grit on the vital 
parts of the projection equipment. 

(a) Should the fire rollers become coated with oil, they will collect dust and 
grit, which will scratch the emulsion on the film. Such marring and destruc- 
tion of film is very costly, and definitely does not contribute to our war effort. 

(&) Deposits of oil, grease, and grit on the film strippers cause wearing of 
the sprockets and damage to the sprocket-holes of the film. 

(c) Excessive oil on take-up devices causes them to slip, resulting in film 
mutilation by pile-up or sprocket breakage. 

( d) All containers of oil should be kept carefully covered, and oil cans should 
be cleaned before being refilled. 

(3) Make Only Necesary Replacements. Due to the difficulty of 
obtaining replacement parts, it will be necessary to make the present 
parts last longer. The projectionist must assume greater responsi- 
bility in his care of the apparatus he operates. This means a daily 
inspection of the various items of the apparatus to insure to the 
utmost degree continuous, efficient operation. To a large extent this 
can be accomplished by systematic care to eliminate abnormal wear. 

The projector mechanism has many precision-made parts. To re- 
duce replacements and repairs to a minimum, the projectionist 
should keep his eyes constantly open for signs of uneven or jerky mo- 
tion of the mechanism, and his ears attuned to any unusual noises 
during operation. A good practice is to turn the projector over by 
hand before the start of each day's show to see whether it revolves 
freely or not. If it seems to bind, the switch must not be thrown or 
serious damage may result. With the projector idle, try by hand the 
meshing of the teeth of the main drive gear, the lower sprocket pinion 
gear, and the intermediate gear. When the teeth on any or all of these 



June, 1942] WARTIME CONSERVATION IN PROJECTION 519 

gears show signs of rapid wear they should be realigned, otherwise 
new gears will shortly have to be installed. 

At least once a week check the synchronizing marks on the vertical 
shaft gear, the intermediate gear, and on the intermittent movement 
flywheel to see whether they are in their proper operating relation. 
Watch the intermittent. Any slack that may develop between the 
star and the cam, or in cam and flywheel shafts, should be removed 
and every visible screw should be tightened at least once a month. 
This will avoid much future trouble. 

Care should be taken when removing the intermittent sprocket, 
movement, or any other delicate part, not to strike the hard surface 
of the mechanism housing, as the good parts may be burred or jarred 
out of perfect alignment. 

When the intermittent sprocket or star-wheel shows undue wear, 
tension on the pad or film guide should be checked and the spring 
compressed or released until the desired tension on both sides of the 
shoe is obtained. Too much tension wears the sprockets and may 
damage the film. 

The pad rollers should be adjusted by the simple method of placing 
two thicknesses of standard 35-mm film on the sprocket held tightly 
over the teeth. The surfaces of the roller should be allowed barely 
to touch the film, and then the arm is tightened in this position. The 
rollers should be in line with the sprocket-teeth; that is, the teeth 
should operate in the recess formed in the rollers. A good practice is 
to wash the sprocket-teeth at least twice a week with a stiff-haired 
brush dipped in kerosene, and at least once a month the entire mecha- 
nism should be thoroughly cleaned with kerosene to remove all in- 
jurious foreign bodies. 

Always, when making repairs, or installing gears, make sure before- 
hand that the proper procedure is thoroughly understood and that 
guide marks are scribed by hand on the parts or that the factory guide 
marks match in order to have perfect alignment. Proper tools should 
be available before starting any such work. 

On some mechanisms the stripper plates and sprockets may be re- 
versed when they show undue wear, but such reversing should be done 
very carefully and after some thought, as in some cases more harm can 
be done than good. 

In the care and maintenance of the sound-head, practically the 
same precautions should be followed as indicated for the upkeep 
of the projector mechanism. The many electrical connections should 



520 WARTIME CONSERVATION IN PROJECTION [J. S. M. P. E. 

be frequently checked and tightened. When a rotary stabilizer is 
used the roller should be left open at all times except when film is 
running in the projector. 

The following list should prove helpful in the care and maintenance 
of the upper and lower magazines : 

Tighten all screws. 

Check the bushings, shafts, and reel locks. 

Watch the upper magazine tension. Excessive tension causes fast wearing of 
the upper feed sprocket. 

Keep the upper friction spring and collars clean and lubricated. Avoid 
jerky upper magazine feed. 

When readjusting the take-ups, place a heavily loaded reel in the lower maga- 
zine. Start the motor, and, beginning with no tension, gradually tighten until 
the reel picks up and revolves slowly from any position in which it is stopped. 
Give an extra half-turn to the adjusting knob and lock it. 

Do not wait for take-up belts to break. Change belts every thirty days, and 
allow oil-soaked leather belts to dry thoroughly. Carefully examine removed 
belts for breaks, bad spots, etc. 

Ventilating fans in rectifiers require periodic inspection and lubri- 
cation from one to two times a year. The rectifier should be located 
in a well ventilated, cool spot. A free flow of air should be main- 
tained. Avoid placing rectifiers too close to other equipment or 
placing materials on top of them. 

In bulb-type rectifiers, the bulb sockets and clips should be in- 
spected to make sure they are clean and not corroded or pitted. 
Sandpaper may be used to remove corrosion in order to make good 
contact. The bulbs should be secure in their sockets, and should be 
checked every few weeks. The various connections should also be 
checked. 

The power input to the rectifier should correspond to the trans- 
former rating. Voltages should be kept as close as possible to the 
recommended values. Variations over 10 per cent should be cor- 
rected. 

A few precautions in the care and maintenance of rewinders, reels, 
splicers, and electrical change-overs will prove helpful in prolonging 
the useful life of the equipment. 

Rewinder alignment should be checked. Aluminum reels should 
be handled with care, as new ones are not available. 

Realign the splicer and check the cutting blades. 

Once a month, check the change-overs and the foot-switches for 
proper contact and alignment. 



June, 1942] WARTIME CONSERVATION IN PROJECTION 521 

(4) Burn Carbons at Minimum Current Specified by Manufacturer. 
Use Carbon Savers Where Available. It is suggested that motion 
picture theaters operate projection lamps at or near the minimum 
arc current recommended for the trim in use if the resulting reduction 
in screen illumination below that at maximum recommended current 
can be safely tolerated. The general adoption of this suggestion 
should result in a considerable power and carbon saving, and for 
those theaters using copper-coated carbons, a substantial reduction 
in copper consumption. 

It is felt that this suggested reduction in operating current, while 
bringing the level of screen illumination below recommended prac- 
tice in many instances, will still permit acceptable projection of 
motion pictures and, for the duration of the war, is justified by the 
substantial saving of power and essential materials which can be ac- 
complished in this manner. 

Check ammeters and voltmeters in projector arc circuits to be cer- 
tain they are accurate, before making any alterations in your present 
operations. 

Check into the availability of reliable carbon savers on the market 
at the present time that will operate satisfactorily in your lamps. 
Be sure to use most economical carbon combination and length of 
carbons available for your lamps. Avoid striking an arc too soon. 

(5) Clean Lenses Properly and Protect Condensers and Reflectors. 
Lenses, condensers, and reflectors should be cleaned with special lens 
tissue or soft cloth. Avoid the use of abrasive cleaning materials or 
cloths containing fibers that scratch. Condensers and reflectors 
should be cleaned only when thoroughly cool, as any sudden cool 
draft may damage the optical system. 

Most arc lamps are equipped with inside protective flame shields. 
These shields should be properly maintained. 

Projection optical systems should be cleaned every day before the 
show. Do not turn the mirror around in its holder, as in a very short 
time the entire surface will be pitted. Do not attempt to remove pits 
forcibly. Check the mirror-retaining clips for the proper holding 
tension; when too tight, the mirror may crack due to expansion. 

Port glasses should be cleaned daily. 

Treated lenses should be cleaned in accordance with the manu- 
facturer's instructions. Keep oil from reaching the lens element. 
These instructions pertain to both sound and projection optical sys- 
tems. Care should be taken to prevent chipping. 



522 WARTIME CONSERVATION IN PROJECTION [j. S. M. P. E. 

(6) Service Regularly All Electric Distribution Points. Motors, 
Generators: Friction is the greatest cause of wear and tear on all rotat- 
ing equipment. Anything that can be done to reduce friction will 
tend to increase the life of all such equipment. It is, therefore, im- 
possible to place too much emphasis on cleanliness of the equipment 
as well as of the surroundings of such equipment. 

In order to prevent dust and dirt from dropping or being blown 
into the unit itself, all walls as well as the floor and ceiling of the 
motor-generator room should be painted and cleaned regularly. 
Lubrication of the unit should be done in accordance with the in- 
structions of the manufacturer and a chart should be kept of such 
lubrication to show the regularity of such service. Bearings should 
be drained at regular intervals of not more than six months and re- 
filled with a good grade of oil, of a viscosity as recommended by the 
manufacturer. 

Brush contact should always be good and the tension should be 
kept at the minimum that will not allow sparking. Brushes should 
be staggered so as to allow even wear across the entire width of the 
commutator. Never use brushes other than the grade recommended 
by the manufacturer. 

Keep all slots in undercut commutators clean by the use of a wooden 
stick of the proper width, and never use oil on any commutator. If 
necessary to use an abrasive on the commutator, clean both brushes 
and commutator thoroughly afterward. Keep the shaft and coup- 
lings in proper alignment. Blow out all dust and dirt from the wind- 
ings of the unit with a blower. 

Alignment of the motor and generator shafts should be checked 
and the couplings kept tight. Misalignment and looseness cause 
vibration, increased wear, and replacements. 

Bus Bars, Fuses, and Switches: These do not ordinarily wear out; 
they generally burn out, and do so because they are not kept clean. 
Dirt or corrosion causes resistance to electric current, and resistance 
causes heating; the heat causes additional resistance and a vicious 
circle is thus built up which eventually destroys the unit. Good 
contact, therefore, is the first requisite of extended life of these units, 
and good contact is maintained by cleanliness and tight connections 
Go over all switchboard connections regularly with proper tools, and 
if refillable-type fuses are used make sure that all -contacts within the 
fuse itself are clean and tight. 

The numerous a-c and d-c switches throughout the projection 



June, 1942] WARTIME CONSERVATION IN PROJECTION 523 

room should be inspected at least once a month. The panels should 
be opened and every nut and bolt, switch, and fuse-holder should be 
cleaned and tightened. 

Resistors: Resistors do not ordinarily wear out if they are properly 
selected for the duty they are to perform. These units need no special 
service except to be kept clean, and all connections to them be kept 
tight. It is suggested that not less than once every six months they 
be freed of all dirt and dust by the use of a blower. Resistors should 
be properly placed and adequately ventilated to prevent overheat- 
ing. 

All outside connections on ballast rheostats should be checked. The 
cover should be removed periodically, and the bolted connections to 
the resistor material checked. Rectifiers should not be operated 
above the recommended rating. Once a year, or oftener if necessary, 
blow out the accumulated dirt and lint in the rectifier stacks. This 
will insure proper ventilation and cooling. 

(7) Allow Sufficient Warming- Up for Vacuum Tubes. Burn Tubes 
at Specified Ratings. It is important that amplifier and rectifier tubes 
be pre-heated and become stabilized at operating temperatures before 
the sound system is operated. Usually a fifteen-minute period is 
sufficient for this purpose. Certain types of tubes, particularly rec- 
tifiers, require a pre-heating period to allow the electron emission to 
become stabilized so that all parts of the filament are liberating elec- 
trons before the plate voltage is applied to the tube. If the plate 
voltage is applied before sufficient electrons have been emitted, the 
surface of the filament may be damaged, or part of the filament may 
be burned away at one spot. 

Many of the larger tubes have spiral extension springs to take up 
the slack of the filament resulting from expansion and elongation due 
to the heating. Sufficient pre-heating time should be allowed to per- 
mit the filament to assume its normal operating position before ap- 
plying the plate voltage. 

Mercury- vapor tubes must be pre-heated to drive the mercury from 
the filament and plate elements of the tube before applying the anode 
voltage. This usually requires three to five minutes, depending upon 
the location of the mercury in the tube and whether or not the tube 
had previously been pre-heated. Tubes of this type should have an 
initial pre-heating period of five to fifteen minutes and then used for 
two or three days. They can then be stored in a vertical position for 
future use. 



524 WARTIME CONSERVATION IN PROJECTION [j. s. M. P. E. 

Once a mercury-vapor tube has had an initial pre-heating, and all 
the mercury has been driven off the tube elements, the daily pre- 
heating period is much shorter than when the tube is first put into 
operation. 

Equipment manufacturers issue instructions regarding pre-heating 
of tubes where necessary. Follow these instructions carefully. 

Tubes should be operated at the voltage ratings specified by the 
manufacturer. An accurate meter should be used in making these 
measurements. A majority of installations are provided with a 110- 
to 120-volt switch to adjust the primary input voltage. If the am- 
plifier or unit is not equipped with a switch of this type, the voltage 
may be adjusted by moving the tap on the primary of the power 
transformer. Operating the tubes on line voltages above the normal 
value will not add anything to the output of the tubes but will only 
decrease their life. Operating the tubes below their normal rated 
voltages also shortens tube life. 

(8) Inspect, Thread, and Rewind Film Carefully. Keep It Clean. 
Film should at all times be carefully handled. It should be kept away 
from all sources of heat, except the normal heat during projection. 
The regulations against smoking should be obeyed. 

Film should under no circumstances be left lying exposed on benches 
or elsewhere, but should be immediately placed in metal containers 
or cabinet after use. 

Film should be inspected each time before it goes through the am- 
chine. The only way that film can be properly inspected is by slowly 
winding the film by hand. Inspection should cover tears, splices, 
and defects in sprocket-holes. Do not use bent reels. Use fresh film 
cement for making all splices. 

Film should be carefully threaded through the machine. It should 
be in proper place on every roller, gate, and sprocket. Excess slack 
at top and bottom of machines should be taken up before the machine 
is started. Magazine doors should be closed as soon as the film is 
threaded and should be kept closed during the entire operation. 

(9) Handle Reels and Film Containers with Care; They Can Not 
Be Replaced. A bracket or rack should be erected on which to keep 
all empty reels instead of allowing them to lie on the floor or elsewhere 
where they may be damaged. 

Film-storage cabinets and shipping cases should be kept clean. 
Bent reels should be saved, as manufacturers are making arrange- 
ments to straighten such reels. 



June, 1942] WARTIME CONSERVATION IN PROJECTION 525 

After putting reels into the film cabinet, the compartment door 
should be closed by hand. It should not be allowed to snap back into 
place by its own weight. Care should be taken that ends of film 
do not stick out. 

(10) Do Not Throw Anything Away. Because of acute shortages 
of many materials, and the difficulty of obtaining replacements for 
theater equipment, all broken and worn out parts should be saved. 

Save all gears made from steel, bronze, brass, or other material. 
Sprockets, pad-rollers, blades and jaws of old switches, copper wire, 
arc-lamp jaws, and other metal parts should be accumulated for dis- 
position at some future date. 

Do not throw away a transformer or motor of any kind. The copper 
can be reclaimed and the cores can be used again. There are some 
manufacturing concerns who will not ship a new transformer unless 
the old one is returned. 

Broken aluminum reels and other aluminum parts should be welded 
or otherwise repaired. This is a critical metal, and if the part can 
not be mended, save the aluminum. Reel and trailer cans should 
be returned to the film exchanges. Nearly every projection room 
has an accumulation of these cans which is taking up valuable space. 
Trailers and sections of film not in use also should be returned to the 
exchanges, who will then forward the film to film reclaiming com- 
panies. 

The country needs copper. Remove and save the copper coating 
from old copper-covered carbon stubs. Save all the copper drippings 
from copper-coated projector carbons. Provide a metal pail in which 
to store the copper drippings and the copper plating. 

Do not throw out used or defective vacuum-tubes of any kind. 
There are many valuable metals used in the manufacture of these 
tubes. Cooperate with the sound engineer in conserving all tubes, 
and other replacement parts. 

Copper wire, rubber-covered cable, and cable plugs will be among 
the articles hard to obtain. Conserve all your available stock. Fix 
up all cables and plugs in use at the present time. Save all the old 
pieces. Provide a box and some space in which to store all defective 
or burned out electric light bulbs. Each bulb is made of material 
that the country needs for manufacturing new products. 

Keep the accumulation of metal parts by placing metals of one kind 
in one box or pail and metals of another kind in another box. This 
will assist in keeping the different metals separated and facilitate in 
disposing of them when instructions for disposal are received. 



THE DEFENSE PROGRAM OF THE MOTION PICTURE 

THEATER* 



HENRY ANDERSON** 

Summary. Civilian defense involves many technical problems in the solution of 
which the highly specialized talent of the SMPE may be of great assistance. Theaters 
in England continue to operate during bombings and it is planned to continue their 
operation in the United States, due to the important part which they play in maintain- 
ing morale. 

New responsibilities have been placed upon theater management, in the preparation 
of their theaters and the training of their staffs to meet the new emergencies. The 
Motion Picture Theater Industry has risen to its new responsibilities. 

Civilian Defense Engineering is the newest branch of Engineering. 
Like all forms it overlaps other branches, but it has a sufficient num- 
ber of unique specialized engineering problems to dignify it as a pro- 
fession. 

In the Society of Motion Picture Engineers there is represented 
some of the most highly specialized, technically trained talent in the 
country. There are leaders in the fields of light, sound, color, physics, 
chemistry, optics, television, radio, who should be able to assist in 
solving many of the civilian defense engineering problems with which 
the motion picture industry may be confronted. The Society has 
offered its services to the industry and to the nation. 

The engineer and the scientist are playing an important part in 
civilian defense activities, such as: 

(a) Selecting and improving places of safety for civilians in case of air raid. 
They have studied the effects of explosive bombs, and the principles of construc- 
tion for bomb resistance. Much technical information has been obtained from 
England, and extensive tests and research have been carried on here. 

(6) Notifying the population of the imminence or occurrence of an air raid. 
This involves the design of alarm systems, sirens, etc. There has been much dis- 
satisfaction with the existing outside sirens and many of the local alarm devices. 

(c) Extinguishing incendiary bombs and fires set by bombs. The fire protec- 
tion engineer has devoted much skill and research to developing devices for ex- 
tinguishing incendiary bombs, and for strengthening public and private fire pro- 
tection. 

* Presented at the 1942 Spring Meeting at Hollywood, Calif. 
** Paramount Pictures, Inc., New York, N. Y. 
526 



THE DEFENSE PROGRAM OF THE THEATER 527 

(d) Regulating traffic to avoid accidents. Here the traffic engineer has come 
forward. 

(e) Restoring public services damaged as a result of raids. This is almost 
entirely an engineering problem involving repair of lighting, power, fire alarm and 
similar services. 

(/) Preparing for and executing blackouts. This is entirely within the province 
of the lighting engineer. 

(g) Protecting against gas; designing gas masks, providing gas shelters; 
decontamination . 

(ft) Demolishing, clearing, and restoring damaged structures, and rescuing 
persons trapped in fallen buildings. 

(i) Effecting camouflage. This is a combined architectural and engineering 
problem, involving also photography. 

Our ingenuity is taxed to the utmost today to accomplish these 
things without undue use or labor and materials, and here the engi- 
neer and scientist come to the front. 

Civilian defense is directed toward the preservation of civil life and 
property, through organization of the civilians themselves. There 
are no sharply denned battlefronts today nor is there a sharply defined 
line between the civilian and the soldier as far as exposure to danger is 
concerned. The soldier risks his life in battle, and the civilian may 
properly be expected to risk his to the extent that military necessity 
requires. 

For complete civilian protection against bombs there would be re- 
quired shelters, sixty to a hundred feet under ground and of sufficient 
capacity to hold the entire civilian population. In England, in the 
early stages of the war, it became obvious that the provision of such 
shelters was impossible. In the United States, with a coastal zone 
subject to bombing 300 miles or more deep and many thousands of 
miles long, to provide such shelters would consume all our productive 
man-hours and strategic materials and we should have to stop build- 
ing planes, tanks, guns, and ships. 

Where astronomical numbers are involved totals run up quickly. 
If a stirrup-pump were manufactured for every home in this zone, due 
to the nature and quantity of materials involved, war production 
would have to stop. This may be true even of the lowly galvanized- 
iron pail. 

In England, the theaters were closed by the authorities when air 
raids first occurred. There followed an alarming falling-off in public 
morale, and the theaters were reopened and have remained open ever 
since. The authorities in this country came to the decision in the 
early part of the war that the continued operation of motion picture 



528 H. ANDERSON [J. s. M. P. E. 

theaters is important. We in the motion picture industry were among 
the first to prepare our properties for defense emergencies, and as a 
result the civilian defense authorities have encouraged us to regulate 
ourselves as an industry. In many cities and States there is a sepa- 
rate motion picture theater group headed by a practical theater 
operator. Of course, in carrying out measures for protection of 
theaters, we ran into the limitations of avoiding interference with the 
defense effort, and of a shortage of many needed materials and equip- 
ment. 

A theater may be one of the safest buildings in the community. 
Due to lack of window and door openings and its generally substan- 
tial construction, it will be highly resistant to bomb fragments. It 
is provided with a heating and ventilating system, seats, drinking 
water, and other facilities contributing to comfort and safety. 

The authorities urge that the public remain in the theater upon the 
occurrence of an air raid or air raid alarm. In fact, theaters are 
generally required to take pedestrians in for shelter in addition to the 
patrons already in the theater. In England the audience remains in 
the theater and the show is continued during air raids. 

Organization for civilian defense will remain for some time a 
definite part of theater operation. Management finds a whole new 
set of regulations under which it must operate. New responsibili- 
ties have been placed upon it. A theater manager may have in his 
care and be responsible for the safety of more persons than is the com- 
mander of a battleship. He knows his problems and how to handle 
theater emergencies better than anyone in the community and should 
make certain on his own account of the safety of theater audiences. 

The first step in the preparation of a theater is cleanliness through- 
out. A theater is much less likely to be fired by a bomb or incendi- 
arism if combustible material in the theater is reduced to a minimum. 

Cleanliness is the first step in any kind of prevention, whether it be 
accident prevention, fire prevention, or otherwise. Cleanliness once 
established should be maintained by constant inspection. 

Following the clean-up should come the painting or white- washing 
of out-of-the-way places, dark passages, alleyways, etc. Dark corners 
where rubbish may accumulate are eliminated. The probability of 
incendiarism and of intruders secreting themselves in the theater is 
lessened. In case of blackout or electric current failure, a white- 
washed or painted exitway is safer than one unpainted. Whitewash 
is an excellent fire retardant. 



June, 1942] THE DEFENSE PROGRAM OP THE THEATER 529 

In an emergency, people need leadership. Without leadership, 
confusion and panic may occur. Most theaters have a well organ- 
ized staff. The staffs are, however, being drilled more intensively 
than ever before in the handling of emergencies. The handling of 
every contingency that can reasonably be anticipated is being re- 
hearsed with the full theater staff cooperating. 

There can be no standard or uniform procedure, as much depends 
upon the construction and layout of the particular theater. The 
manager, in each case, is called upon to exercise his ingenuity in de- 
vising a routine for his particular house. As many employes as 
possible should obtain the benefits of an air raid warden's course. 

We in the theater business know by experience a great deal about 
handling people and about crowd psychology. Studies have been 
made in England of the reaction of persons in air raid shelters to fear 
and shock. Practical psychologists have classified the types of re- 
action and have given the remedies. For example, some persons 
become hysterical, some freeze in place, some can not possibly remain 
seated, and each must be handled differently to avoid precipitating 
panic. Further study should be made of this important and inter- 
esting subject to the great advantage of the industry. This type of 
information might be made available to theater staffs to assist them 
in handling emergencies. 

Theater management has a definite moral and legal duty to notify 
the audience immediately upon the occurrence of an air raid alarm. 
Physicians, air raid wardens, nurses, and others in the theater have 
duties elsewhere and must be given opportunity to leave the theater. 

The house lights should be turned on, and notice should be given 
by the responsible head of the theater on duty at the time. He 
should recommend that the audience remain seated and avoid the 
greater danger of the streets. The show that has been momentarily 
interrupted should then go on. The engineer can assist in the design 
and layout of public-address or other systems for giving prompt and 
definite notice to the audience. 

A theater presents a comparatively simple black-out problem, for 
it has few windows and doors. The principal problem is the main 
entrance. Here, by using subdued lighting near the entrance, little 
light should reach the street. Here we believe the lighting engineers 
in this Society can be of assistance in establishing standards for the 
type, arrangement, and color of lamps and the degree of illumination. 

Fire escapes and alleyways should be sufficiently illuminated to 



530 H. ANDERSON [j. s. M. P. E. 

provide a safe means of exit. The Society may be helpful in finding 
or designing an acceptable outside exit light and establishing accept- 
able standards of illumination. 

There should be provided an emergency lighting system supplied 
by batteries. Shortages and priorities rule out the internal combus- 
tion engine as a source of current supply. Each theater is a special 
lighting problem, and we have the situation today of each manager 
working out a lighting system for his theater as best he may. There 
is much help that the lighting engineer can give in selecting and 
standardizing equipment. 

Luminous materials, including paints, are being advocated for 
marking exits and for parts of ushers uniforms. Some of these are 
permanently luminous; others require periodic exposure to light; 
others require ultraviolet light to activate them. None of them 
should be adopted before testing under actual conditions. Here are 
opportunities for the Society to establish standards for these mate- 
rials. 

The fire-protection engineer has given much intelligent study to 
methods of extinguishing incendiary bombs, and certain definite 
standards for materials and equipment have been established. In 
spite of this there is a host of salesmen attempting to foist upon the 
public trick extinguishers, powders, etc. We must remember that 
the problem is not so much to extinguish the bomb itself, but to 
control and extinguish the fire started by the bomb. 

Sabotage has, in the opinion of qualified persons, evidenced itself 
very little as yet. We know of none attempted in theaters. The 
operator of any property should, however, be informed on this im- 
portant subject. The forms aimed at theaters might be : 

(1} Incendiarism, causing fires. 
(2) The planting of bombs in public places. 

(5) Creating disturbances or alarm in order to instill fear and make people 
jittery. 

(4) Miscellaneous damage. 

Enemy agents are provided with ingenious and effective devices 
for starting incendiary fires. The devices may be designed to operate 
several hours after having been planted. There is little personal 
danger in handling or extinguishing devices of this type. A clean 
theater and frequent inspections are the best preventives. 

Should a bomb be discovered it should not be disturbed, for it may 
contain a mechanism designed to set it off if it is lifted or moved. 



June, 1942] THE DEFENSE PROGRAM OF THE THEATER 531 

Nor should the bomb be plunged in water. Bombs may be designed 
to operate upon being submerged in water. Inasmuch as the bomb 
must not be disturbed, the theater must be emptied at once upon dis- 
covery of it, and the authorities should be notified immediately. 

Each manager should quietly discuss the matter of sabotage and 
the handling of bombs with the appointed authorities, whether they 
be the bomb squad, the police, the fire department, or the F. B. I. 
Where there is no regularly appointed authority an explosives expert 
may be available. 

Sandbags, felt mattresses, or other protective devices should be 
used only by experts, for attempting to confine the explosion may 
only intensify it. 

The primary concern of the theater operator is the preservation of 
human life. He is interested only secondarily in the preservation of 
his property. 

If the manager suspects that any disturbances in the theater may 
be planned sabotage, he should inform the local police or the F. B. I., 
avoiding, however, any liability for false arrest, defamation of char- 
acter, etc. Sabotage may take the form of damage to the theater or 
equipment, such as motion picture machines, sound equipment, or 
switchboards. 

No one should be admitted to remote portions of the theater, 
particularly the stage, dressing rooms, and basement. A thorough 
search for hidden intruders should be made upon closing. Doors 
should be provided with modern locks. Access to the theater by 
means of fire escapes and roof should be prevented. Night crews 
should report any unusual occurrences. Police should try doors on 
their night rounds. 

Inasmuch as the civilian defense publications relating to theaters 
have not discussed sabotage we have covered the subject here in some 
detail. 

In conclusion, this is no time for hysteria nor is it a time to blind 
ourselves to grim realities, whatever they may be. We in the mo- 
tion picture industry have proved many times in the past our full 
sense of responsibility to the community and we shall continue to do 
so with renewed energy and determination. 



TECHNICAL PROGRESS IN THE MOTION PICTURE INDUS- 
TRY OF THE SOVIET UNION* 



G. L. IRSKY** 



Summary. Ten years ago the young Soviet motion picture industry was merely 
in the embryonic stage and had no basic technical facilities to cope with the problems 
inherent in the new industry. 

At present there are factories in the Soviet Union producing equipment, films, and 
accessories for the motion picture studios and theaters. 

In recent years many studios have been rebuilt and adapted for sound motion pic- 
tures and a number of technical problems connected with it were solved. As for the 
near future, there are improvements to be made in the quality of raw films as well as 
further improvement in the printing and developing technic. 



Industry in its entirety is young in the Soviet Union, but the 
motion picture industry is the youngest of all industrial branches. 
We have not yet, indeed, reached the high standards of the American 
motion picture technic. However, if we consider the accomplish- 
ments of the last ten years, the significant progress of this young 
industry will become evident. 

With the rapid development of the sound motion picture it became 
evident to us ten years ago that satisfactory results could hardly be 
expected with what means we possessed at that time. Our studios 
were not adapted for producing sound-films, nor were the accoustics 
of our theaters adequate to meet the requirements of the sound 
picture. Furthermore, our laboratories did not have the necessary- 
developing and printing equipment so that most of this work was 
done by hand. 

The difficulties in making sound pictures were further aggravated 
by the lack of enterprises to manufacture noiseless cameras, special 
types of microphones, amplifiers, and sound-recording equipment. 
The distribution of the available sound pictures was limited by the 
shortage in the necessary projection apparatus. 



* Presented at the 1942 Spring Meeting at Hollywood, Calif. 
** Chief Engineer of the Motion Picture Industry of the Soviet Union. 

532 



MOTION PICTURES IN THE SOVIET UNION 533 

We had only one plant producing films, and this one could not 
guarantee the quality of its product since the technological processes 
used there were old and outdated. Furthermore, the facilities of 
the plant were very inadequate to meet the demand for positive and 
negative films. 

We lacked also the necessary experience and personnel. The aver- 
age age of our motion picture engineers, at the present time, is be- 
tween 24 and 30 years. Hence, ten years ago there were no people 
who could undertake the solution of the multitudinous problems in- 
volved in the new development. 

For the elimination of all the above-mentioned difficulties, time 
was required. As yet we have not overcome them all. However, 
the work done has laid the foundation for a speedy development of 
our motion picture industry. Unfortunately, the war has to some 
extent impeded the conversion of a whole series of very vital and al- 
ready finished experiments into practical use. But I emphasize 
"to some extent," since our motion picture industry as well as the 
entire industrial structure of the Soviet Union is rapidly overcoming 
the effects of destruction brought about by our enemy's temporary 
successes, and with the enthusiastic and self-sacrificing assistance of 
our people, continues to progress. 

In this report I shall endeavor to describe the basic factors char- 
acterizing the progress of our motion picture industry. In the past 
ten years, we have built a number of factories in Moscow, Leningrad, 
and other cities for the manufacture of various kinds of motion pic- 
ture equipment. Our inventors and designers have built a few new 
types of newsreel cameras. These cameras were manufactured by us 
in our factories, and the results proved to be most satisfactory under 
various operating conditions. It is well to mention that among these 
newsreel cameras the one with the inside magazine has received high 
praise from the cameramen. 

Another camera noteworthy is one used in trick cinematog- 
raphy, with a very stable mechanism assuring accurate positioning 
of the picture frame. A few groups of such cameras have been built 
and have given satisfactory results. In 1940 our factories manu- 
factured the first types of noiseless sound cameras. 

A few years ago our factories began to manufacture automatic 
developing machines, and as a result, most of our studios switched 
from hand developing to machine procedure. Our factories make 
several types of developing equipment: some are small machines for 



534 G. L. IRSKY [j. S. M. P. E. 

studio laboratories and others are of considerable capacity for estab- 
lishments supplying prints to the motion picture theaters. 

In our country we aim to increase the development of national 
culture, which naturally is reflected in the cinematographic art. 
More than eighty languages are spoken by the vast populace of the 
Soviet Union. A great number do not understand the Russian lan- 
guage. These films must be accessible to the entire population, and 
therefore eighty per cent of our pictures are re-recorded into an aver- 
age number of 30 to 40 national languages by means of dubbing. 

The national actors are invited to view the pictures, then they 
repeat the words in their own languages before the microphone. 
Difficulties were encountered in this phase of the work, for it is not an 
easy task to control the time elements involved in synchronizing the 
voice with the allotted sound-track. 

We have designed special equipment for controlling the synchro- 
nism of the dubbed voice and the action on the screen. We have made 
great progress in this branch of cinematography, and are proud to 
have overcome the many obstacles in our way. 

There are at the present time, in the Soviet Union, about 40,000 
motion picture installations equipped with domestic projection 
machines. Our movie theaters are of various seating capacities and 
are, therefore, furnished with various types of sound projection 
equipment. Some are stationary installations equipped with high- 
intensity arc lamps; others are medium-size installations, and some 
equipment is of the portable 35-mm and 16-mm types mounted on 
vehicles. 

For the past few years, our factories have been making sound- 
recording equipment designed by the Russian Professors A. Shorin 
and P. Tager. Four years ago we started also to manufacture the 
RCA sound-recording systems. We have had some difficulties in 
making a precision-type sound-recording galvanometer. Some of 
the difficulties have been eliminated by our own efforts, and we hope 
to overcome others with the assistance of the RCA. 

The domestic supply of motion picture equipment and films is 
insufficient to meet the demand, so that in addition to the foreign 
purchases of equipment, most of it coming from the United States, 
we are endeavoring to increase our own production. 

(1) Improvements in Sound Recording. Our industry set-up is 
quite different from what you have over here, with the concentration 
of the majority of the studios in Hollywood. Under the Soviet 



June, 1942] MOTION PICTURES IN THE SOVIET UNION 535 

Union conditions, where great attention is being paid to the cultural 
requirements of the various nationalities of our land, we must have 
separate studios in all the national republics of the USSR. Most of 
these sixteen republics already have such studios. The servicing of 
these studios, spread over great distances, is naturally difficult. This, 
however, does not interfere with their development. On the con- 
trary, due to communication difficulties they have become more inde- 
pendent. At the same time the experiences of each studio are shared 
by all the others. 

We were greatly concerned with the accoustical improvement of 
our studios. We have achieved optimum reverberation in most of 
our leading studios, such as in the Mosfilm, Lenfilm, Detfilm, and 
Georgian studios. The new Mosfilm studio meets all the require- 
ments necessary for the production of first-rate sound recordings. 

Considerable work has been done in sound-proofing our studios 
from external noises and vibration. 

In 1938 we started to produce the RCA equipment PM-38 and to 
date a few dozen units have been made. As mentioned heretofore, 
we had some difficulties with the RCA mirror galvanometers, which 
were also experienced previously in the Hollywood studios. In order 
to eliminate these difficulties a noise-reduction bilateral shutter is 
being substituted for the three-angle mask, and to improve the 
speech recording, which came out previously with considerable am- 
plitude distortion, a compressor was put into use. 

To obtain greater amplifier stability in sound recording and 
reproduction as well as to improve the frequency response, and reduce 
interference and distortion, we are using to a great extent the prin- 
ciple of feed-back amplification. 

We are not satisfied altogether, as yet, with the quality of our 
sound recordings. The reasons for this are to be found in the low 
standard of our sound-recording films and in the shortcomings of the 
galvanometers,- essentially in their amplitude characteristics. 

So far as the galvanometers are concerned, we have indeed made 
considerable improvements. We are still experiencing production 
difficulties and the amount of rejections is quite high. We hope to 
solve this problem too. In case of failure, we shall be compelled to 
adopt a new modulator made by Professor Shorin, which is easy to 
produce and assures a high quality of sound recording, but which will 
necessitate considerable changes in the technological processes. 

Recently we have paid considerable attention to the problems of 



536 G. L. IRSKY [j. S. M. P. E. 

re-recording and multi-channel recording, in improving the electro- 
acoustical characteristics of the sound-recording channels. We have 
begun the production of new microphones, but the quantities are yet 
insufficient to meet the demand. A limited quantity of testing and 
measuring equipment produced by us has enabled us to standardize 
the sound-recording procedures and controls. 

The results of these improvements made can readily be seen in 
such pictures as Moscow Laughs, Circus, A Musical Story and Anton 
Ivanovitch. 

(2) Improvements in the Film. The quality of the raw film is very 
essential in determining the final sound and picture quality. As 
mentioned before, we had one old plant for manufacturing raw films. 
While we were successful in obtaining satisfactory results under semi- 
laboratory conditions, and a limited supply of good-quality film, the 
same could not apply after we went into mass production of film. 
However, before mass production of film could be started, consider- 
able research work had to be done on the development of the new 
technological processes. A number of very interesting processes 
have been developed, including new processes of ripening and wash- 
ing the emulsions, drying the films, making multi-layer emulsions, 
etc. 

These new processes have not yet been applied under production 
conditions, but will be applied in the new plant recently built. This 
new plant was completed during the war, and is now producing films 
for civilian consumption as well as for the army. After the war we 
plan to produce films in sufficient quantities to cover our demands. 

(3) Improvements in Printing and Developing. The improvements 
in the printing and developing processes could not be made while 
these processes were manual. The introduction of automatic me- 
chanical equipment resulted in some improvement and we hope still 
further to improve this situation in the very near future. We have 
already taken steps in this direction. 

A year ago we made the first line of printing machines, and finished 
a machine for non-slip sound-track printing based upon RCA speci- 
fications. We had some difficulties with this machine, since the non- 
slip printing mechanism did not have the necessary operational sta- 
bility. 

We have developed standard procedures for printing and develop- 
ing and have produced some control and test equipment. The 
quantity of this equipment is far from meeting our demands so that 



June, 1942] MOTION PICTURES IN THE SOVIET UNION 537 

we have had to purchase and consider purchasing more printing 
machines in the United States. 

Last year we were trying to improve the developing process. Some 
experimental work has recently been completed on the effect of 
turbulation of developers, and we hope that this and other research 
will enable us in the near future to improve considerably our technics 
of printing and developing. 

(4) Color- Films. Up to recently our color-films were limited to 
two-color processes. We have well equipped laboratories at the 
Moscow and Kiev studios for two-color films. Some of our color 
productions, as, for instance, Grunia Kornakova, The Fair of Soro- 
chino, and others, have been very well received. 

In view of the fact that the two-color films made it impossible to 
utilize fully the color potentialities, a few years ago we started 
experiments with three-color processes. The main work in this 
connection is carried on in the Lenfilm and Mosfilm studios and by 
the Scientific Research Institute. 

Due to lack of experience, we have encountered a number of 
difficulties, especially in carrying over the experimental work from 
the laboratory to the studio. We are working on several methods in 
order to find the best one. 

The best results have so far been obtained with a special camera 
using three-color films. Eighty per cent of our cartoons are made in 
color, and we believe that the color cartoon is the first step in the 
further development of our color pictures. 

It is our opinion from our limited experience in the production of 
color-films, and in spite of satisfactory results obtained by using 
three films, this method will not be acceptable on account of the 
bulkiness of the camera and the complexity of the printing and 
developing processes. 

We believe that real progress in color pictures will be attained only 
after means are devised whereby color pictures may be made with 
the standard camera, using a single film, and after developing arid 
printing procedures are simplified. This is more easily said then 
done, and we well realize the difficulties that will be encountered. 
We shall find it necessary to devise new methods for developing and 
printing color-films and to rebuild some departments of the film 
manufacturing factories. We will have to solve our technical prob- 
lems more rapidly than heretofore and we intend to obtain the assis- 



538 G. L. IRSKY ft. S. M. P. E. 

tance of American specialists in the field, as we have done in regard to 
sound recording. 

(5) Stereo-Motion Pictures. In the Soviet Union, as in many other 
countries, we have experimented with the so-called "three-dimen- 
sional movies." We should like to see our movies as representing 
real life, and are not satisfied with the flat two-dimensional scenes. 
After experimenting with all the known methods of screen viewing 
with special eyeglasses, as, for instance, the anaglyph method, the 
polarizing method, eyeglasses with shutters, etc., we have come to the 
conclusion that none of the methods based upon the use of eyeglasses 
could have any practical application. We would rather equip our 
theaters than our audiences. 

The inventor, Mr. S. Ivanov, has built a screen consisting of two 
surfaces: one surface (rear) is a standard motion picture screen, but 
the other (front) consists of a frame with metal wires spreading fan- 
wise from the lower part of the screen. The picture is taken with a 
standard camera equipped with a device that divides the image into 
two, so that an object is photographed from two separate points, just 
as each of our eyes records a separate and different view of an object. 
During the projection of the film each eye of the spectator sees a 
separate image, and a depth effect is obtained. 

A special motion picture theater has been built in Moscow for show- 
ing stereoscopic films without eyeglasses. The impression upon the 
audience is so strong that the spectators forget they are in a theater. 
Imagine "seeing" pigeons flying about almost in the middle of the 
hall, or cigarette smoke spreading over one's head! 

We attach considerable importance to stereoscopic motion pictures 
without the use of eyeglasses. The work done thus far is only the 
beginning, and we believe that Mr. Ivanov's method has a bright 
future. However, the work is only in the experimental stage and re- 
quires some development to simplify the screen, increase the bril- 
liancy, etc. We cherish the idea of combining the stereoscopic images 
on the screen with stereophonic sound, based upon the developments 
of RCA and Bell Telephone Laboratories. 

(6) Improvements in Projection. Ten years ago we had about 
20,000 installations, of which only 2 or 3 per cent were sound; but 
now we have close to 40,000 motion picture installations and 80 per 
cent are sound. They are of many varieties. In addition to regular 
motion picture theaters we have also a tremendous number of port- 
able equipments. 



June, 1942] MOTION PICTURES IN THE SOVIET UNION 539 

We pay considerable attention to the cultural requirements of our 
villages. Whatever is done in our cities is immediately carried out to 
our villages. Many of the latter are located far from railroad lines 
so that the equipment together with the films must be transported on 
carts, trucks, and special automobiles. For some localities having no 
electricity, power equipment must be transported. 

Each village and collective farm desires its own motion picture 
installation, and it is quite difficult to satisfy the demand for equip- 
ment and films. In order to meet the demand we have a great num- 
ber of portable equipments for transporting both equipment and 
films from place to place. 

There are some places in the Soviet Union, for example, the Cau- 
casian Mountains, which during certain periods (in winter and early 
spring) are not accessible either by horses or automobile. To such 
places the men and equipment are carried in planes. 

Our Red Army likes the movies, too. For them we provide special 
vehicles equipped with motion picture installations, radios, micro- 
phones, and phonographs. These vehicles also carry special equip- 
ment for showing pictures in the military camps during the daytime 
without the need for darkened places. 

The average number of prints made from each new film is from 
500 to 1000, and often this amount is not sufficient. 

Such a great requirement for prints has made it necessary to adopt 
measures for prolonging the life of the films. We have developed 
special treatments for the care of prints, making it possible to increase 
their runs from eight to ten times. 

The amplifying and acoustical equipment of our theaters previously 
was of obsolete types which made it difficult to obtain high-quality 
reproduction. Two years ago we designed some modern amplifiers 
which are now being installed in the theaters. 

One of these amplifiers is of particular interest: Professor P. 
Timofeev of Moscow and Engineer B. Kubetzki of Leningrad have 
designed a secondary-emission phototube multiplier, which has made 
possible the building of a very compact amplifier. The electronic 
beam of Professor Timofeev's phototube is focused electrostatically. 
Its sensitivity is about 500 milliamperes per lumen and the noise level 
is around minus 55 to 60 db; it has an absolutely flat frequency re- 
sponse curve for the audio-frequency range; total voltage applied, 
500 volts. 

This photo-multiplier is substituted for the phototube and pre- 



540 G. L. IRSKY 

amplifier. Depending upon the size of the theater, a power output 
stage is connected to it, usually consisting of two tubes (push-pull 
circuit) . It is a light, compact amplifier easy to service. 

Another important development consists of a high-pressure 120- 
volt a-c or d-c gas lamp having a rating of 250 to 300 watts. The gas 
pressure in operation is 80 to 100 atmospheres. The brilliancy of 
this lamp is equivalent to a 3-kw arc lamp. Its average life is 200 
hours. 

We have begun also to produce new loud speakers with permanent 
magnets, selenic rectifiers, motion picture screens, etc. 

CONCLUSION 

In addition to the problems already mentioned, on which we have 
been working for the past few years, there are many more to solve in 
the very near future. The major and immediate problems to be 
solved by our institutes, laboratories, and factories are quality im- 
provements in sound recording and further development of color and 
stereoscopic pictures. We hope that you in American industry will 
continue to help us as you have done in the past. We are confident 
that, with the growing number of experienced personnel, and the help 
of the now powerful Soviet Union industry, we shall speedily over- 
come these problems. 

I feel that we are justified in such expectations in the light of the 
achievements accomplished during the recent years by our people. 



THE DEVELOPMENT OF THE SOUND-FILM 



JOHN E. ABBOTT** 

Summary. With the appearance of the part-talking film "The Jazz Singer" in 
1927 a new era began. Talking films had been introduced before, but not until then 
had the engineers solved the problem of amplifying as well as of recording and re- 
producing sound. As often, technical invention preceded creative use and at first the 
new machines were used clumsily. Yet it is well to recall the character of certain 
primitive talkies in order to understand what the elements were which stirred the 
wonderment and curiosity of the public, and it is the only effective way of appreciating 
the speed and ingenuity with which this apparent retrogression in film-making was 
overcome during the next few years. 



The profound interdependence of film technique and the technics 
of film is nowhere seen so clearly as in those first years of the transi- 
tion to sound 1927 to 1932. Silent pictures had developed a method 
of telling a story all their own: by the end of the silent era, the film 
medium had been enriched by the work of creative directors, had 
been made a fluent instrument for the expression, through pantomime 
and suggestion, of even the most complicated emotions and ideas. 
Only with the utmost reluctance did the lovers of cinema surrender 
to that blasting, squalling but terribly popular new plaything, the 
talkies. To them it seemed that motion pictures, in gaining a voice, 
had lost a soul. 

And actually it had lost almost everything that was at the root of 
cinematic creation. Freedom, speed, the penetrating close-up, the 
broad, revealing pan and all outdoors simply disappeared overnight 
as the microphone overwhelmed the studios in the great onrush of 
sound. The inclination, of course, is to say that the bulky s